literature review on heat transfer fluids and thermal...
TRANSCRIPT
Literature Review on Heat Transfer Fluidsand Thermal Energy Storage Systems in
CSP PlantsSTERG Report
Lukas Heller∗
May 31, 2013
Contents
1. Introduction 41.1. Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
I. Heat Transfer Fluids 5
2. Qualities of HTFs 5
3. Liquids 53.1. Synthetic Oil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53.2. Molten Salts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63.3. Liquid Metals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.3.1. Sodium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93.3.2. NaK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103.3.3. LBE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113.3.4. Summary Liquid Metals . . . . . . . . . . . . . . . . . . . . . . . . 11
3.4. Liquid Glasses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
4. Gases 114.1. Air . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124.2. Other Gases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144.3. Summary Gaseous HTFs . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
∗Lukas [email protected] - Solar Thermal Energy Research Group, Stellenbosch University
1
5. Solids 145.1. Particle Receiver . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
6. Fluids with Phase Change 156.1. Direct Steam Generation (DSG) . . . . . . . . . . . . . . . . . . . . . . . 15
7. Supercritical Fluids 167.1. s-H2O . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167.2. s-CO2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
II. Thermal Energy Storage Systems 22
8. Introduction 22
9. Media for Active Sensible Heat TESSs 249.1. Storage Media in Use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249.2. Molten Salts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259.3. Sodium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259.4. Haloglass™ RX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269.5. Steam/Water . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
10.Media for Passive Sensible TESS/Filler Material 2710.1. Packed Beds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2710.2. Rocks and Sand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2810.3. Concrete . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3010.4. Ceramics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3110.5. Graphite . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
11.Latent Heat Storage Media 3211.1. Salts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3311.2. Metals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
12.Chemical Potential Storage Media 35
13.System Considerations 36
14.Modeling of TESS 37
References 40
Appendices 55
A. Properties of HTFs 56A.1. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56A.2. Synthetic Oil Therminol® VP-1/Dowtherm® A . . . . . . . . . . . . . . 59
2
A.3. Solar Salt™ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60A.4. HITEC® Heat Transfer Salt . . . . . . . . . . . . . . . . . . . . . . . . . 61A.5. Low Temperature Salt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62A.6. High Temperature Salts . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63A.7. Sodium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65A.8. Eutectic Sodium-Potassium Alloy . . . . . . . . . . . . . . . . . . . . . . . 66A.9. Eutectic Lead-Bismuth Alloy . . . . . . . . . . . . . . . . . . . . . . . . . 67A.10.Ambient Air . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68A.11.Pressurized Air . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69A.12.DSG - Superheated . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70A.13.Supercritical H2O . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71A.14.Supercritical CO2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
B. Storage Media 73B.1. Potential Sensible Heat Storage Media . . . . . . . . . . . . . . . . . . . . 73
B.1.1. Haloglass™ RX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77B.2. PCMs with Tmelt > 550 ◦C . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
3
Nomenclature
CR(S) central receiver (system)
CSP concentrating solar power
DNI direct normal irradiation
DSG direct steam generation
HTF heat transfer fluid
LCOE levelized cost of electricity
LHTESS latent heat thermal energy storage system
PCM phase change material
TES(S) thermal energy storage (system)
1. Introduction
Concentrating Solar Power (CSP) plants generate electricity by concentrating direct nor-mal irradiation (DNI) from the sun through mirror systems on receivers and convertingthe gained thermal energy into electricity in a heat engine. The main advantage of CSPas compared to other renewable energy technologies (like, for example, wind power andphotovoltaics) is the possibility of relatively cheap and efficient storability and thereforedispatchability of energy. This is due to the conversion of the incoming radiation intothermal energy, which can be stored cost-efficiently.
On the other hand, this conversion means that CSP plants have thermodynamic cyclesthat limit their efficiency by the high and low temperature of the Carnot cycle. In orderto achieve high conversion efficiencies, high temperatures at the heat engine energy sourceand low temperatures at its energy sink are necessary. These temperatures are limitedby the heat transfer fluid (HTF) in the receiver and the working fluid in the heat enginecycle (which can be one and the same). Storage options have to be chosen according tothe desired temperature range(s) of the HTF and the properties of the working fluid.
1.1. Scope
The scope of this review is to give an overview on research which has been done on HTFsfor CSP plants and on media being utilized in thermal energy storage systems (TESS).The focus hereby is on high-efficiency/high-temperature cycles with large thermal energystorage systems—and therefore central receiver systems (CRS).
4
Part I.
Heat Transfer Fluids
2. Qualities of HTFs
HTFs can be classified by their states of matter during normal operating conditions.Additionally to the three standard states (gaseous, liquid, solid), HTFs that undergo aphase change and supercritical fluids are also possible.
Becker (1980) rated potential HTFs for CSP applications by their thermal and trans-port properties. After a first assessment, he focused on a commercial molten salt(HITEC® Heat Transfer Salt), a commercial heat transfer oil, air, hydrogen, helium, wa-ter vapor, sodium, potassium, mercury and ammonia. Cabeza et al. (2012) summarizedthe state of the art and the conducted research mostly on TESS for CSP applications.However, Cabeza et al. also looked at the heat transfer to and from the storage systemand, therefore, the available heat transfer fluids.
Important thermophysical properties of HTFs are:
• low lower temperature limitation (solidification temperature)
• high upper temperature limitation (evaporation temperature/thermal stability limit)at low pressures
• high thermal conductivity → receiver temperature close to HTF temperature
• low viscosity → lower pumping power requirements
• high density and heat capacity → enable use as storage medium
• possibility of usage as working fluid
• chemical compatibility (low corrosivity) with contact materials
• low cost, high availability
• low toxicity, flammability, explosivity and environmental hazard
3. Liquids
3.1. Synthetic Oil
Almost all commercial parabolic trough CSP plants to date use oil as the heat transferfluid (NREL, 2013a). In most cases this is either Therminol® VP-1 or Dowtherm® Asynthetic oils. These, however, limit the upper operating temperature to approximately400 ◦C (Dow Chemical Company, 2001; Solutia Inc., 2013). Other disadvantages of theoils are degradation over time, high cost and inflammability. In the following, syntheticoils are not considered because their temperature limitations prevent high efficiencycycles.
5
3.2. Molten Salts
The first CSP pilot plants that used liquid salt as the HTF and thermal storage mediumwere the 1 MWe Molten-Salt Electric Experiment (MSEE ), the 2.5 MWe THEMIS andthe 10 MWe Solar Two central receiver power plants (Reilly and Kolb, 2001; Dunnet al., 2012). The operating temperature range of the latter was 290 ◦C to 565 ◦C with abinary salt, composed of 60 % of NaNO3 and 40 % of KNO3 (by mass) - so called SolarSalt™. Although Solar Salt™ is stable to higher temperatures of up to 600 ◦C, this lowermaximum temperature has been chosen because its corrosion rate with the used stainlesssteel is acceptable at that temperature (Pacheco et al., 2000). The difference betweenthe chosen lower temperature and the solidification temperature at 222 ◦C is to establisha safety margin for freezing.
The design steam parameters were 535 ◦C/100 bar in the 35 MWt steam-generator/-superheater unit and 510 ◦C/100 bar in the condenser turbine. The latter was refurbishedfrom the Solar One predecessor and therefore a limiting factor in plant efficiency (Tyneret al., 1995; Pacheco et al., 2000). Pacheco et al. and Litwin (2002) summarized theresults gained from tests and operation between 1996 and 1999. From these, Zavoico(2001) and Moore et al. (2010) deducted design options and standards for future centralreceiver (CR) molten salt power plants in great detail. On this basis, the next plantof this type was built - arguably the most advanced CSP plant to date: The 19.9 MWe
Solar Tres/Gemasolar plant with 15 full-load hours of molten salt thermal energy storage(TES) (Lata et al., 2008). The latter enables 24 h power generation on summer daysand, therefore, baseload capability. The nominal turbine inlet temperature generated inthe steam generator is 542 ◦C (SIEMENS AG, 2010). Operating experience of the firstyear of power production of the plant are summarized by Garcıa and Calvo (2012).
Heat transfer characteristics of molten salts are mediocre. The reasonably high den-sity and mediocre specific heat capacity enable a low volume flow but the low thermalconductivity leads to elevated temperatures on the outside of the receiver pipes and,therefore, high radiation losses. Rodrıguez-Sanchez et al. (2013) investigated the influ-ence of the number and diameter of receiver tubes in a Gemasolar -like plant on maximumtube temperature, maximum molten salt film temperature, HTF pressure drop and re-ceiver cost (see Figure 1). The heat transfer between pipe and HTF can be improvedby increasing the fluid velocity and turbulence (for example, by usage of spiral tubes asshown by Yang et al., 2010). Another way of improving a salt receiver’s efficiency is toimprove its optical efficiency, as done by Garbrecht et al. (2012) who built pyramid-likespikes in which the HTF circulates. The pyramids act as a trap for light and thermalradiation. The biggest advantage of molten salt as the HTF is the possibility of directstorage at relatively low costs.
Advanced Molten Salts The high solidification temperatures of liquid salts are prob-lematic especially in line-focusing CSP plants (parabolic trough or linear Fresnel re-ceivers) because the HTF would freeze during the night or in times of low irradiation.In CR plants, the salt will normally be drained into a tank while filling the receiverwith gas. However, salt freezing, for instance, due to blocked valves, can still occur and
6
Figure 1: Temperatures over flow length through molten salt receiver (Lt: tube length;Nl: Number of lines in the receiver or salt paths; Np: Number of panels in thereceiver) (Rodrıguez-Sanchez et al., 2013).
cause failures (Pacheco and Dunkin, 1996). Other possible solutions besides draininginclude trace heating or circulation of stored hot salts but all of these would result inhigher heat losses, electrical power consumption and/or investment costs. To evade thiscomplication, research is being done on liquid salts with lower melting points.
Raade and Padowitz (2011) reported the experimental finding of a quinary moltensalt composition with a melting temperature as low as 65 ◦C and thermal stability above500 ◦C. However, they state that the cost of the found salt “[. . . ] is likely to be con-siderably higher than the simple binary Solar Salt [. . . ].” Bauer et al. (2012) report theinstallation of a testing loop for degradation, stratification and corrosion testing of newsalt compositions with melting temperatures as low as 75 ◦C and thermal stability com-parable to Solar Salt™. They found enhanced thermal stability of the salts under oxygenenriched air. Siegel et al. (2011) measured the thermophysical properties of differentlow-melting-point molten salts.
Research is also currently being done on molten salts with higher maximum operatingtemperatures in order to allow higher efficiency power cycles (see for example, U.S. De-partment of Energy, 2012a). Kelly (2010) of Abengoa Solar wrote an extensive reporton possible future implementation of supercritical Rankine cycles in CSP power plants.He discussed several scenarios at live steam temperatures of 565 ◦C, 590 ◦C and 650 ◦C(subcritical, supercritical and ultra-supercritical Rankine cycle, respectively) with dif-
7
ferent HTF/working fluid/TES combinations, two of which using molten salt as theHTF. Kolb (2011) estimated the economic benefits of raising the salt HTF’s receiverexit temperature to approximately the same temperatures as investigated by Kelly andpredicted levelized cost of electricity (LCOE) reductions of up to 8 %.
Raade et al. (2012) found a quinary composition of LiCl, NaCl, KCl, CsCl and SrCl2with a melting point of 253 ◦C at ambient pressure and thermal stability up to ap-proximately 750 ◦C. The proposed maximum operating temperature of the so-calledSaltstream™ 700 is 700 ◦C. The novel salt consists of more than 70 % by weight of CsCland LiCl, which are both expensive materials. The aim is to find compositions with re-duced shares of these substances without considerable penalty on the thermal qualities.
Williams (2006) assessed different salts for use as coolants in next generation nuclearpower plants. These allow for operation well above 700 ◦C at low pressure, however, theyalso mostly have high solidification temperatures (above 300 ◦C) and are substantiallymore expensive than, for example, Solar Salt™. Corrosion issues also have to be investi-gated for a lack of operating experience. Forsberg et al. (2007) proposed some of thesesalts for use in solar power towers with direct thermocline TESS and graphite as thefiller material. The chosen power cycle in their model is a closed multi-reheat Braytoncycle with helium or nitrogen as the working fluid and operating temperatures between700 ◦C and 1000 ◦C.
Olson et al. (2009) and Sabharwall et al. (2010) conducted experiments on a corrosiontest loop at the University of Wisconsin - Madison. They heated up two different moltensalts that could potentially be used in next generation nuclear power plants up to 500 ◦C,namely so-called FLiNaK, which consists of LiF-NaF-KF (46.5-11.5-42 mole percentage),and KCl-MgCl2 (67-33 mole percentage), and observed the material loss of the followingpipe materials over several hundreds of hours: Hastelloy N, Hastelloy X, Inconel 617,Haynes 230 and Incoloy 800H. Sabharwall et al. found that the used graphite capsulesgreatly enhanced the corrosion rate. Because of this, the found values are much higherthan in a graphite-free environment and shouldn’t be used for deciding on an appropriatecomponent material.
3.3. Liquid Metals
Other HTFs don’t have the problematic upper and lower operating T emperature limita-tions of molten salt. For example, liquid metals and their alloys can have solidificationtemperatures below 0 ◦C and boiling temperatures above 1600 ◦C. Freezing of the HTFinside pipes, the receiver, valves and TESS can, therefore, practically be eliminated. Atthe same time, the HTF can operate at low pressures and still reach the temperaturesrequired for a next-generation Rankine or a Brayton power cycle.
Another upside of liquid metals are the outstanding heat transfer characteristics andthe low viscosity. Due to the high thermal conductivity, the temperature gradient ofthe flow inside the receiver pipe will be very small. Additionally, the pipe thicknesscan be kept small as well because of the low pressure. This leads to maximum pipetemperatures close to the fluid’s exit temperatures (Boerema et al., 2012) - resultingin higher receiver efficiencies - and reduced strain inside the pipe caused by thermal
8
expansion (Lata et al., 2008). Eventually, these properties allow higher maximum solarfluxes on the receiver and, thus, a higher thermal efficiency of it. Kelly (2010) statedthe maximum allowable incident flux on a molten salt receiver (20 bar inside pressure,696 ◦C maximum outer pipe temperature) to be as high as 2.5 MWt. This is notablyhigher than fluxes in realized projects, which, according to Lata et al., are 0.8 MWt forSolar Two and 1.0 MWt for Gemasolar. Liquid metal receivers can tolerate even higherinfluxes.
Pacio and Wetzel (2013) assessed different types of liquid metals as HTFs for CRSs.They investigated sodium, the eutectic lead bismuth composition and tin as candidatematerials and stated their advantages, limitations and areas of recommended futureresearch.
3.3.1. Sodium
So far, the focus in liquid metal HTF research was on sodium. In the 1980s’ Small SolarPower Systems Project of the International Energy Agency (IEA-SSPS), a sodium-cooledexternal receiver was tested at the Plataforma Solar de Almerıa (Schiel and Geyer, 1988).Its nominal incoming power was 2.7 MWt at a maximum fluid temperature of 560 ◦C anda maximum heat flux of 1.4 MWt/m2. However, the tube receiver was also tested at aradiation input of up to 3.4 MWt with a heat flux of up to 2.5 MWt/m2, producing pipetemperatures up to 770 ◦C. Post-experimental metallurgical analyses didn’t show anysignificant deformations or creep damage of the pipes even under these super-nominalconditions.
Boerema et al. (2011) showed with a high-level comparison of HITEC® Heat TransferSalt and sodium as HTFs for CR plants, that the main advantage of the latter (besidesthe higher operating temperature) is the lower pipe temperatures due to the high thermalconductivity. This enables higher radiation fluxes (as mentioned above), smaller apertureareas and, therefore, lower heat losses (see Figure 2). However, the heliostat field has tobe able to focus on this smaller target and stresses on the pipes are potentially higher.
A big concern with the use of sodium as an HTF is the strong exothermic reaction withwater in which hydrogen is one of the products. In 1986, the IEA-SSPS project endedin a sodium fire, destroying some of the equipment. The International Atomic EnergyAgency (1999) reported the fire was a sodium spray fire that resulted in sodium spillageof approximately 10 000 kg and burned at 225 ◦C, while An (2011) and Boerema et al.(2012) mentioned 14 000 kg of leakage and a maximum flame temperature of 1200 ◦C.The incident was caused by maintenance procedures on a valve. This event stopped mostresearch on sodium in CSP but the nuclear industry has continued working on sodiumas a nuclear reactor coolant (for example, Poplavskii et al., 2004). Guidez et al. (2008)stated that the combined reactor time of sodium-cooled plants exceeds 388 years withmostly promising experiences gained. However, there have been sodium leaks and oftenfires in almost all reactors that ever went critical (Poplavskii et al., 2004).
Recently, liquid metals, including sodium, have been investigated as HTFs for CSPplants again (Singer et al., 2010; Kotze, Backstrom and Erens, 2012b; Boerema et al.,2012; U.S. Department of Energy, 2012b; Hering et al., 2012; Pacio and Wetzel, 2013).
9
850
900
950
1000
1050
1100
1150
1200
1250
1300
Pip
egsu
rfac
egte
mpe
ratu
rega
tgexi
tg
2 3 4 5 676
78
80
82
84
86
88
90
92
Effi
cien
cyg(
o)
PipegLengthg(m)
Hitec:gEfficiencySodium:gEfficiencyHitec:gSurfacegTSodium:gsurfacegT
Figure 2: HITEC® Heat Transfer Salt and sodium receiver efficiencies and maximumsurface temperatures for various pipe lengths. Emissivity = 0.85, concentrationratio = 1300, D = 9 mm (Boerema et al., 2012).
Hering et al. described the possible direct conversion of the heat in the sodium to electricpower in an alkali metal thermal electric converter (AMTEC) cycle. The possible useof magnetohydrodynamic (MHD) pumps for sodium could be of merit because of theirlow maintenance requirements. The AMTEC technology is not available for the givenparameters yet.
3.3.2. NaK
The eutectic sodium-potassium alloy NaK78 (22.2-77.8 mass percentage) melts at am-bient pressure at −12.6 ◦C and boils at 785 ◦C (Foust, 1972). Despite the inferior heattransfer characteristics of NaK78 as compared to sodium, the low solidification pointmakes it very attractive for transient power plants, like CSP. Freezing issues in pipes,vents and the receiver are practically eliminated. Other non-eutectic NaK alloys couldshow more favorable thermodynamic properties for plants, like higher densities, at thecost of higher solidification temperatures (see Kotze, Backstrom and Erens, 2012b). Oth-erwise, the characteristics of and issues with NaK are very similar to those of pure sodiumand can be found in the elaborate handbook by Foust.
Diver et al. (1990) presented the state of the art in parabolic dish CSP systems atthe beginning of the 1990s. They focus on indirectly heated Stirling engines with liquidmetal HTF that are evaporated in the receiver and condense on the heat exchanger tothe engine. They mention sodium, potassium and NaK78 as investigated and tested
10
HTFs, depending on temperature limits in the receiver.
3.3.3. LBE
Another liquid metal alloy investigated for CSP applications is the lead bismuth eutecticcomposition Pb-Bi (44.5-55.5) (LBE). According to Pacio and Wetzel (2013), it does nothave the problematic drawback of sodium or NaK (reaction with water) and has a veryhigh density, leading to much lower flow speed requirements. The boiling temperatureof LBE (1670 ◦C) is even higher than that of sodium but the solidification temperature(125 ◦C) is higher as well, so that freezing is an issue. Additionally, the high corrosivityand cost of LBE could be problematic (Furukawa et al., 2004; Zhang and Li, 2008; Pacioand Wetzel, 2013).
3.3.4. Summary Liquid Metals
Liquid Metals have very high potential as HTFs because of their wide range of practicaloperating temperatures and the superior heat transfer characteristics. They allow forhigh maximum fluid temperatures at low pressures, high receiver efficiencies and lowpressure drops. However, they don’t qualify as a direct storage medium due to theirhigh costs and are challenging in terms of operation, maintenance, safety and steelcorrosion.
3.4. Liquid Glasses
Glasses are known to be chemically stable and withstand high temperatures. However,they have high melting temperatures and high viscosities even at elevated temperaturesso that pumping becomes problematic. Halotechnics (2013) introduced Haloglass™ RX,a glass which is pumpable down to 450 ◦C and has some properties that would qualifyit as an HTF. Due to the high minimum operating temperature it is not promoted herebut instead presented as a possible high-temperature storage medium in Section 9.4 withsome more properties given in APPENDIX B.1.1.
4. Gases
The upper temperatures of gaseous HTFs in CSP systems are usually only limited bythe materials of the receiver pipes, ducts, etc. They are therefore especially suited forhigh-temperature applications. Gases, however, have the downside of low heat transfercoefficients and densities.
Gases can be used to directly power a gas turbine, thus making use of the very hightemperatures which can be generated in a CR. The exhaust gases can be used to powera bottoming (Rankine) cycle, which renders possible high thermal system efficiencies.
11
4.1. Air
Air is the obvious choice as the HTF of a Brayton cycle. It is available everywhere,non-hazardous, theoretically free of cost and does not necessitate a heat exchanger forco-firing. On the other hand, it has a low density and unfavorable heat transfer char-acteristics and therefore makes big heat exchangers and receivers necessary. For theimplementation of a CSP receiver in a gas turbine (GT), it has to be pressurized. Thepressure drop has to be paid special attention to because the GT’s performance is highlydependent on the pressure drop in between compressor and expander.
Receivers are the crucial element of CSP plants with air as the HTF. Due to thedifficult heat transfer characteristics, effort has to be made to reduce the aperture areaand temperature differences between the receiver and the fluid.
In the SOLGATE project, several institutes and companies (including the DLR1 andCIEMAT2) developed and tested a 250 kWe CR CSP prototype at the Plataforma Solarde Almerıa (EC, 2005). It consisted of three air-cooled receiver modules connected inseries. The low temperature (LT), intermediate temperature (IT) and high tempera-ture (HT) receiver modules heated up the pressurized air approximately from 300 ◦C to550 ◦C, to 730 ◦C and to 960 ◦C, respectively. The nominal expander inlet temperatureof the GT was raised to about 1200 ◦C by a gas combustor. This co-firing enhanced theGT efficiency while decreasing the solar share of the thermal energy input.
The SOLHYCO project was the successor of SOLGATE. It featured a 100 kWe micro-turbine with recuperator and a combustor for bio-diesel (DLR, 2010). The recuperatornoticeably increases the inlet temperature into the receiver and, therefore, enables asingle high-temperature receiver instead of the threefold receiver system in SOLGATE.The HT receiver was changed considerably from a pressurized volumetric to a pressur-ized tube design featuring novel profiled multilayer (PML) tubes. These address twoproblems that lead to high temperature differences between the outside of the pipe andthe final receiver outlet temperature: a) the high temperature gradient between irradi-ated and non-irradiated side of the tubes is decreased by introducing a layer of copper inbetween two thin concentric steel pipes, which increases the thermal conductivity of theresulting pipe; and b) the heat transfer to the air flow is increased by adding a wire-coilstructure to the inner pipe wall (see Figure 3). The PML tubes were not installed intothe final receiver because of manufacturing delays, however, laboratory tests with themshowed the expected homogenization of the tube temperatures. For the use in a commer-cial power plant, some issues with the intermetallic connection’s durability would haveto be resolved. The maximum receiver outlet temperature reached was approximately800 ◦C. Several design flaws, that limited this temperature and the receiver efficiency,could be identified and solutions proposed.
The next stage in the development of a solar combined cycle power plant was theincrease of the size to a demonstration plant with a 4.6 MWe industrial gas turbine(without bottoming cycle). The Solugas Consortium (2012) consists of Abengoa Solar,the DLR, GEA, Turbomach and New Energy Algeria. The start-up tests of the plant
1DLR: German Aerospace Center2CIEMAT: Center for Energy, Environment and Technological Research
12
Figure 3: Visualization of different tube designs (EC, 2005).
were commenced in May 2012 but no detailed information could be retrieved.Schwarzbozl et al. (2006) economically analyzed different designs for hybridized so-
lar gas turbine prototype systems between 1 MWe and 15 MWe for two different loca-tions. They found LCOE values for solar-generated energy between 0.13 EUR/kW he
and 0.90 EUR/kW he depending on unit size and solar share.Other receiver designs were proposed, for example, a pressurized closed volumetric
receiver by Hischier et al. (2009), an irradiated ceramic plate heat exchanger by Jenschet al. (2012) and others, as summarized by Avila Marın (2011). An Israeli Company,Aora Solar Ltd (2012), makes use of a solar-driven microturbine to generate off-gridpower and heat (with considerable co-firing of natural gas).
Air can also be used as an HTF without being the working fluid. In this case thehot air can drive, for example, a Rankine steam cycle through a heat exchanger. Forthis application, an open air receiver is usually used for simplicity reasons. The airdoesn’t have to be pressurized (except to overcome pressure drops in receiver, piping andheat exchanger) and the turbine is not directly coupled with the receiver outlet flow.However, the advantage of using a high temperature working fluid can not be made use of,since Rankine cycles are today limited to approximately 640 ◦C. The only commissioneddemonstration plant using this technology for grid-power is the Julich Power Tower, asdescribed by Hennecke et al. (2009). In the AlSol project, this technology is planned tobe used in a 7.1 MWe hybridized solar-natural gas power tower (Koll et al., 2011).
Wilson Solarpower (2010a) proposed a system in which unpressurized air is used asthe HTF in a closed receiver. This hot air then heats pressurized air in a regenerativeheat exchanger (see Figure 4) to be used in the microturbine of a Brayton cycle.
13
Figure 4: Regenerative air-air heat exchanger (Wilson Solarpower, 2010b).
4.2. Other Gases
Other Gases, like helium, CO2 or nitrogen could also be used as HTFs for their superiorheat transfer and flow characteristics or material compatibility (corrosion) as comparedto air (Becker, 1980). Massidda and Varone (2007), for example, analyzed heat trans-fer, pipe stresses and pressure drops for helium as the HTF in absorber tubes. Theyalso investigated heat transfer enhancing measures, like swirl tapes or increased piperoughness. All of these gases have specific problems and are far from the demonstrationphase.
4.3. Summary Gaseous HTFs
Air is by far the most investigated gaseous HTF. This is because of its practically infi-nite availability and extensive experience with it as a heat transfer and working fluid.The high operating temperatures of the fluid enable combined cycle plants with highefficiencies, however, the heat transfer poses a problem due to the HTFs’ low densityand thermal conductivity.
5. Solids
5.1. Particle Receiver
Wu et al. (2011) compared three different direct absorption receivers (DARs), two ofwhich were open particle receivers. In DARs the heat is directly absorbed in the form ofradiation by the HTF (see Figure 5) instead of being transferred through, for example,steal pipes in the form of conduction and then being transferred to the HTF via convec-tion. The maximum receiver temperature can, therefore, be found in the HTF and noton the receiver material which leads to decreased radiation losses and material stress.Open particle receivers feature solid particles surrounded by a fluid. This can have sev-eral advantages, for example, increased radiation absorption, higher heat capacity orlower material temperatures towards the surroundings.
Chen et al. (2007) developed a computational fluid dynamics (CFD) model of an openparticle receiver in form of a curtain of ceramic particles between 200 µm and 600 µm insize. Their simulations show good agreement with experimental results. The calculated
14
Fig. 1 Schematic illustration of aerodynamic and thermal pro-cesses in a solid-particle receiver
Figure 5: Schematic illustration of aerodynamic and thermal processes in a solid-particlereceiver (Chen et al., 2007).
receiver efficiencies for the small and simple receiver are below 70 % for particle outlettemperatures of less than 1000 K.
Crocker and Miller (2011) modeled a cylindrical volumetric receiver with air andcarbon nano particles as the HTF. Their initial CFD simulations suggest fluid outlettemperatures of up to 1430 K but many questions of the design weren’t answered at theearly stage of the research.
Thermal efficiencies of solar particle receivers are expected to reach 90 % (Ho, 2010).However, they are still in an experimental stage and several questions, for example, theheat transfer to the working fluid, are to be answered.
6. Fluids with Phase Change
6.1. Direct Steam Generation (DSG)
The direct generation of steam inside the receiver (DSG) has been the subject to researchand development for a long time. The Solar One tower as well as the first two commercialpower tower plants, PS10 and PS20 with a power rating of 10 MWe and 20 MWe use(d)saturated steam as the HTF (NREL, 2013a). The latter both produce steam at 40 barand 45 bar, respectively, at an outlet temperature below 300 ◦C and have been runningfor several years.
The first two-large scale linear Fresnel solar power plants, PE1 (1.4 MWe) and PE2(30 MWe), rely on the same HTF but evaporate the water in a line focusing system(Novatec Solar, 2013). In nominal conditions, the steam in the receiver pipes is heated
15
to 270 ◦C at a pressure of 55 bar, however, temperatures above 500 ◦C have been achievedduring tests in the PE1 plant.
DSG is difficult to realize in parabolic trough power plants because of the relativelyhigh pressures necessary and the resulting stresses on joints. However, numerous stud-ies investigated the technology and see high potential for cost reduction and efficiencyincrease. These have been summarized by Birnbaum et al. (2008). Due to the two-phase flow inside the pipes, the heat transfer is difficult to exactly measure or predict.Therefore, more water is added to the evaporator to ensure sufficient heat transfer and awater-steam-separator has to be incorporated to protect the turbine from high moisturecontent (Mertins, 2009).
Eck and Zarza (2006) compared saturated with superheated steam parabolic troughDSG plants’ performances. They found that the benefits in power block efficiency of asuperheated plant is often compensated for by the higher thermal receiver losses. Thedecision for one of the two designs has to be made under consideration of part-loadbehavior, TES implementation and cost (investment as well as operation and mainte-nance).
Feldhoff et al. (2012) compared different steam generation modes in parabolic troughDSG plants and further investigated the once through method.
The TES poses a big problem in DSG plants. Because of the incorporation of preheat-ing, evaporation and superheating, a combination of sensible heat and a latent heat TESseems most viable. Due to the temperature gradient between HTF and the latent TES,the steam pressure during storage discharge has to be significantly lower than when theturbine is fed directly from the solar field (Birnbaum et al., 2008). More information onthe implementation of TESs in DSG plants can be found in Section 13.
7. Supercritical Fluids
7.1. s-H2O
Supercritical water (s-H2O, pcrit,H20 = 221 bar) has been used as a working fluid in con-ventional power plants for decades. The state of the art are so called ‘Ultra-Supercritical’(USC) steam plants with parameters of up to 620 ◦C/310 bar (Boss et al., 2007). USCturbines are usually installed in big coal power plants, their ratings range between ap-proximately 200 MWe and 1050 MWe and the plants reach net efficiencies in excess of45 %. Next generation advanced USC (A-USC) plants are expected to run at turbineinlet temperatures of between 700 ◦C and 760 ◦C (Weitzel, 2011) with efficiencies above50 %.
Due to the high critical pressure of water, s-H2O requires special - usually expensive -materials for piping, turbomachinery and heat exchangers. No solar power plant usings-H2O as the HTF has been built so far. However, Coventry and Pye (2010) proposed aparabolic dish system, employing s-H2O or superheated (subcritical) water as the HTFand working fluid with a molten salt system as the storage. The advantage of s-H2O isthe lack of the evaporation process which leads to smoother isobaric heating processesas compared to superheated fluids. The heat transfer between the fluid and another
16
single phase fluid (for example, molten salt) is, therefore, exergetically preferrable (seeFigure 6). Coventry and Pye found a slightly higher overall efficiency for their USC cycleas compared to the superheated cycle. However, the outcome highly depends on systemparameters and assumptions.
Figure 6: Subcritical and supercritical heat transfer in systems with one HTF (Coventryand Pye, 2010).
7.2. s-CO2
Compared to water, carbon dioxide has a much lower critical pressure of 73 bar, yet su-percritical CO2 (s-CO2) is relatively dense at approximately 0.6 kg/m3 (Wright, Conboyand Rochau, 2011). On the one hand, these characteristics decrease stresses on pipesand flow speeds, on the other hand, turbomachinery and heat exchangers at a fractionof the size of steam components can be realized. Turchi (2009) gave an introduction tos-CO2 as an HTF and/or working fluid in CSP plants and listed (dis-)advantages whencompared to other candidate HTFs. Gary et al. (2011) presented an s-CO2 combinedcycle solar power tower with a LCOE of 0.06 USD/kW he as one of the targets for 2020in the SunShot initiative.
Sandia National Laboratories have done extensive research on s-CO2 cycles for severaldifferent applications (solar, geothermal, nuclear) and summarized the testing and devel-opment outcome in (Wright, Conboy, Parma, Tom G. Lewis, Gary A. Rochau and AhtiJ. Suo-Anttila, 2011) and (Wright, Conboy and Rochau, 2011). They proposed cyclesin the small to medium range (0.1–10 MWe), for instance, in modular CR systems.
Chapman and Arias (2009) compared three parabolic trough configurations with syn-thetic oil, subcritical CO2 and s-CO2 as HTFs, respectively. They found that the pump-ing work for subcritical loops would be orders of magnitude higher than in the base case(oil). Thanks to its higher density, s-CO2 is much more favorable in that sense. However,the high pressure dramatically increases the requirements on pipes and will most likelynot be feasible in line-focus systems but only in CR systems.
Chapman and Arias (2009) as well as Ma and Turchi (2011) addressed the problemof adding a TESS to a s-CO2 system. The latter proposed molten salt as the storage
17
medium, however, this would limit the turbine inlet temperature (at least during dis-charging) to much less than 600 ◦C. Active direct and passive storage systems appearnot viable due to the need for high pressure tanks.
Cox (2009) found that standard heat transfer correlations are in general applicable tos-CO2 heat exchangers. However, in close proximity to the critical point the measuredheat transfer noticeably differed from the calculated values.
Rouillard et al. (2009) compared the corrosion effects on a ferritic-martensitic anddifferent austenitic steels in contact with s-CO2 at a temperature of 550 ◦C and a pressureof 250 bar. Their research was aimed at the development of components for the secondarycycle for advanced (for example, sodium-cooled) nuclear reactors. It turned out thatcorrosion had a much more profound effect on ferritic-martensitic steels as compared toaustenitic ones.
Characteristics as Working Fluid Dostal et al. (2004) gave an elaborate overview on s-CO2 as a working fluid and possibly HTF in nuclear power plants. The information foundin their report is also applicable to CSP, as they investigate different configurations, forexample, with liquid metal cooled reactors and s-CO2 only as the working fluid. In thisdesign, one advantage in terms of safety is that there is no direct heat exchanger betweenliquid metal and a water/steam cycle. CO2 also reacts exothermally with liquid metalsbut no hydrogen is created in the reaction, therefore greatly decreasing the hazardsassociated with liquid metals. Dostal et al. built a model for different s-CO2 cycles - someof them with preheating, reheating, precooling or intercooling - and compared them witheach other and to a supercritical water and an ideal gas Brayton cycle with helium as theworking fluid. With this model, they investigated the influence of heat exchanger (pre-cooler, regenerator, etc.) sizes on thermal efficiency and pressure losses. In general, theysee great potential in s-CO2 cycles due to the simple layout of Brayton cycles as comparedto Rankine cycles and the high efficiencies at moderate pressures/temperatures. Thelatter is caused by the cycles’ low compression work due to the low compressibility factorof CO2 at the critical point. Mentioned problems are little experience in compressorsthat work in close proximity of the critical point, higher corrosion rates than those ofhelium and recuperators for real gases (pinch-point problem).
Kato et al. (2004) investigated the implications of precooling and intercooling in CO2
power cycles at pressure levels of 70 bar and 125 bar for nuclear reactors with a fluid outlettemperature of 800 ◦C. In their model, certain configurations of compressor stages anda bypass compressor improved cycle efficiency by up to 6 percentage points.
Muto and Kato (2007) compared s-CO2 cycles at different temperatures (500–650 ◦C),pressures (80–200 bar) and with single or dual expansion for next generation nuclearreactors in the power rating range of 200 MWe to 600 MWe. The found efficiencies variedbetween 42.6 % and 50.3 % for the designs with different recuperation concepts. The dualturbine layout allows for a potentially beneficial concept: The highest temperature isproduced in the low pressure turbine stage and its outlet temperature is still higher thanthe inlet temperature to the high pressure stage. The whole heating process for the highpressure side can, therefore, be achieved by an internal recuperating heat exchanger (see
18
Figure 7). The heat source (reactor/solar receiver/etc.) does not get in contact with thehigh pressure fluid, which reduces stresses on the heat generating component.
Figure 7: Scheme and T-s diagramm of dual-expansion s-CO2 power cycle (Muto andKato, 2007)
Another thermodynamic comparison of CO2 cycles was conducted by Kulhanek andDostal (2011). They investigated four different designs — namely the simple Brayton,precompression, recompression and the partial cooling cycle — at turbine inlet tem-peratures of 500 ◦C to 850 ◦C. They found that the partial cooling cycle provides thebest efficiencies, however, at the cost of a more complex design with three compressorunits. At high temperatures, the precompression cycle is the second most efficient andit appears to have favorable part-load properties.
Moisseytsev and Sienicki (2010) conducted s-CO2 power cycle simulations for the VeryHigh Temperature Reactor (VHTR) concept. This type of nuclear reactor is cooled byhelium, which is heated from 400 ◦C to 850 ◦C in the core, producing approximately600 MWt. The high temperature difference poses problems on an s-CO2 power cyclebecause the optimal value for turbines operating between 200 bar and 75 bar is only150 K. Moisseytsev and Sienicki analyzed several solutions to efficiently couple the twocycles. The first one is a cascaded system of three single cycles (Figure 8) with theoptimal specifications stated above. They each have the same heat input of 200 MWt
but their efficiencies vary in a wide range with values of 54 %, 50 % and 44 %, respectively.To avoid a high number of turbomachinery and heat exchangers, the optimal tempera-
ture difference of a s-CO2 cycle can be increased by increasing the turbine pressure ratio.Since increasing the cycle’s maximum pressure creates material problems, a decrease inthe turbine outlet pressure seems more feasible. However, a large portion of the effi-ciency advantages of a s-CO2 cycle is created by compression close to the critical point.Therefore, a cycle with one or two subcritical pre-compression stages and pre-coolers
19
was analyzed. The latter (see Figure 9) has a turbine outlet pressure of approximately11 bar, increasing the temperature difference in the cycle to 430 K. The resulting over-all cycle efficiency (50 %) is higher than that of the cascaded cycle (45 %) while havingsimplified the cycle considerably.
Summary s-CO2 Supercritical CO2 cycles show higher efficiencies than state of the artsteam or air cycles. Other benefits are a higher density, much smaller machinery, simplerplant design and a lower supercritical pressure. These don’t only apply to CSP plantsand much research has been done to promote s-CO2 as a working fluid/HTF in nextgeneration nuclear and conventional power plants as well as for cooling applications.However, considerable effort is still necessary to develop components for utility scalepower ratings (Fuller and Batton, 2009; Sienicki et al., 2011). One of the programsfinanced by the SunShot initiative is the development of a 10 MWe s-CO2 turbine forapplication in CSP plants at Sandia National Laboratories.
20
VHTRBS-CO2 CYCLEBTEMPERATURES,BPRESSURES,BHEATBBALANCE,BANDBEFFICIENCIES
829S5 145.2 108.3l9S9l
CO2 642S6 689So839S9 kg)s 689So l9S96 7S7o6
9lS4B 7S7o785oSo 829S57Sooo l9S9l
22oS3 243S29oSo l83So l9S97 7S67o
He 7S62l l9S97
256S8 9oSlB 89Sl 237S8 243S2kg)s 7S538 9oS3B l9S97 7S669
7ooSo642S6l9S96
32S8 84S4 84S4 9oSl7S62l 2oSoo 2oSoo 7S622
3lS257S4oo 9oS6B
EfficiencyM= 54Sl7 B 5B
32S6 9oSo7S62l 7S62l
3lBQ,MW TL
oC 2Looo 3oSo 4oS3
Input PLMPa 0 kg)s oSlo3 oSlol
95SlB
86.3
151.3
441.3
95S7B
17.1
200
69B
Cycle
96S6B
16.3
TURBINE
HTR
RHXCOMPSMU2
COOLER
COMPSMUl
EffS =
LTREffS =
LTR
EffS =
EffS =
ε XTSC =
ε XTSC =
ε XTSC =
533S2 132.5 87.6l9S9l
CO2 38oS3 4l8S3lo64S3 kg)s 4l8S3 l9S96 7S7o7
92S3B 7S7o855oSo 533S27Sooo l9S9l
l78S3 l9oSo87S9 l8oSl l9S97 7S67l
He 7S622 l9S97
256S8 9oSlB 87So l77S5 l9oSokg)s 7S538 9lSoB l9S97 7S67o
4ooSo38oS3l9S96
32S8 84S3 84S3 87S97S62l 2oSoo 2oSoo 7S622
3lS257S4oo 9lSoB
EfficiencyM= 43S8l B 5B
32S6 87S97S62l 7S62l
3lBQ,MW TL
oC 2Looo 3oSo 42S8
Input PLMPa 0 kg)s oSlo4 oSlol
200
69B
Cycle
96S6B
20.5
95SlB
107.2
130.9
276.2
95S5B
21.3
TURBINE
HTR
RHXCOMPSMU2
COOLER
COMPSMUl
EffS =
LTREffS =
LTR
EffS =
EffS =
ε XTSC =
ε XTSC =
ε XTSC =
68lS2 139.8 99.5l9S9l
CO2 5loS5 553S7937S6 kg)s 553S7 l9S96 7S7o6
9lS9B 7S7o77ooSo 68lS27Sooo l9S9l
l97S9 2l5S388S8 l8lS4 l9S97 7S67o
He 7S62l l9S97
256S8 9oSlB 87S9 2o5S5 2l5S3kg)s 7S538 9oS6B l9S97 7S67o
55oSo5loS5l9S96
32S8 84S4 84S4 88S87S62l 2oSoo 2oSoo 7S622
3lS257S4oo 9oS8B
EfficiencyM= 49S74 B 5B
32S6 88S87S62l 7S62l
3lBQ,MW TL
oC 2Looo 3oSo 4lS4
Input PLMPa 0 kg)s oSlo3 oSlol
200
69B
Cycle
96S6B
18.1
95SlB
95.3
141.0
366.9
95S6B
18.9
TURBINE
HTR
RHXCOMPSMU2
COOLER
COMPSMUl
EffS =
LTREffS =
LTR
EffS =
EffS =
ε XTSC =
ε XTSC =
ε XTSC =
Figure 8: Cascaded s-CO2 power cycle (Moisseytsev and Sienicki, 2010)
21
VHTRNS-CO2 CYCLENTEMPERATURES,NPRESSURES,NHEATNBALANCE,NANDNEFFICIENCIES
827S4 503.1 297.0l9S9l
CO2 397S5 447Sllll3S9 kg)s 447Sl l9S96 lSl27
93SlB lSl3685oSo 827S47Sooo l9S9l
l88S7 2olS494S7 l92S5 l9S97 lSo9l
He 7S439 l9S97
256S8 99S5B 94S5 l86So 2olS4kg)s 7S4l9 88SlB l9S97 lSo86 57.9
4ooSo 32S86397S5 lSo3ol9S96
3lS5 84S7 84S7 88S77S438 2oSoo 2oSoo lSo39
l33S953lS25 3S5277S4oo 89S2B
EfficiencyM= 49S53 B 5B 123.7
3lS4 94S7 3lS777S438 7S438 58B 3S526
Q,MW TLoC loLooo 3oSo 32S4
Input PLMPa 0.21 kg)s oSl2o oSlol 94S77S438
94.7
47.5
600
Cycle
96S6B
18.8
42B
95SlB
101.5
122.6
295.7
97S8B
33.4
TURBINE
HTR
RHXCOMPSMU2
COOLER
COMPSMUl
EffSM=
LTREffSM=
LTR
EffSM=
EffSM=
ε XTSCM=
ε XTSCM=
ε XTSCM=
COMPSMU3
PC2
COMPSMU4
PCl
Figure 9: s-CO2 power cycle with two precoolers (Moisseytsev and Sienicki, 2010)
Part II.
Thermal Energy Storage Systems
8. Introduction
In CSP plants, thermal energy storage systems (TESS) serve multiple purposes. Theybalance the plant in transient periods, for example, during overcast, they enable stableturbine conditions and more full-load hours. The most important reason for the imple-mentation of big TESS is, however, to be able to supply dispatchable or base-load powerto the grid and even stabilize it on demand. This also increases the capacity factor ofthe power block and minimizes defocusing of mirrors.
This section gives an overview of the available types of TESS for CSP applicationsand their characteristics. Alternative introductions to TESS concepts, storage media andtheir heat transfer characteristics can be found in the literature (for example, Cabezaet al., 2012; Gil et al., 2010; Medrano et al., 2010; Li et al., 2011). The letter alsodescribes the modeling of TESSs. As summarized by Duffie and Beckman (1991) andGil et al., the major requirements on TESSs for CSP are:
• (volumetric) energy capacity
• charge and discharge heat rates
• maximum and minimum temperatures, sensible or latent heat storage
22
• safety and environmental impact
• thermal and chemical stability for thousands of cycles in contact with differentmaterials
• heat losses
• quality of the thermocline after charging
• degradation of the thermocline during idle mode
• power requirements for charge and discharge
• costs (for the whole storage system)
NREL (2013a) shows which TES systems have been and are being built into CSPplants. The simplest way of storing heat in a CSP plant is to use the primary HTF asthe storage medium as well, a so-called active direct storage system (see Figure 10(a)).This works well when synthetic oil (as in Luz Industry’s SEGS 3 I plant) or molten salts(as in the Gemasolar plant) are used because they remain liquid at elevated temperaturesbut the high price, especially of synthetic oil, proves a big financial drawback. In DSGplants or when gas is the primary HTF, either a gaseous medium would have to bestored, resulting in low volumetric energy capacities, or a high pressure inside the tankis necessary to condense the medium, which makes long-time storage nonviable.
coldtank
hottank
hottank
coldtank
Figure 10: TESS types: (a) active direct, (b) active indirect, (c) passive
In active indirect storage systems (see Figure 10(b)), the storage medium is sepa-rated from the primary HTF by a heat exchanger. The advantage of this configuration isthat no compromises have to be made in finding a medium that serves as both, an HTFand the storage medium. Instead two ‘specialized’ media can be used. The trade-offis between the avoided losses in the heat transfer/component cost reduction on the onehand and media cost reduction/efficiency improvement through the use of optimizedmedia on the other hand.3SEGS: Solar Energy Generation System
23
So far, commercial TESSs are all of the active two-tank type, with approximately thevolume of storage medium in the system to fill one of them. One attempt to lower thecost of TESSs is to omit one tank and use one-tank for the hot and the cold storagemedium. The separation between the states can be achieved, for example, by use ofa moving separation disk (see Hering et al., 2012) or thermocline separation due todensity differences in the medium. Thermocline systems can be fitted with solid so-called“filler material” with high thermal capacity. Compared to active one-tank systems, fillermaterial can enhance the overall capacity, improve the thermocline quality and savecosts by substituting an expensive fluid storage medium with cheaper solid particles, forexample, ceramics or rocks. Storage systems in which the filler material material is themain contributor to the capacity are also referred to as ‘passive’ (see Figure 10(c)).Depending on the used solid material and HTF, the materials can be in direct contactor separated (for example, by pipes, meshes or capsules around the storage medium toavoid chemical/mechanical interactions or improve heat conductivity).
Forms of Storing Thermal Energy According to Gil et al. (2010), thermal energy canbe stored in three different forms: as a temperature rise (sensible heat), a phase change(latent heat) or chemical potential. Chemical and phase change material (PCM) storagesystems promise great opportunities but are still subject to research whereas all existingTESS of CSP plants store sensible heat.
Previous Reviews Pilkington Solar International GmbH (2000) and Herrmann andKearney (2002) gave elaborate overviews of storage systems and media investigated andbuilt in the 1980s and 90s and some information on promising PCMs and solid media forpassive TESS. Gil et al. (2010) gave an updated review on TESS for CSP applications.Kelly (2010) investigated the technical feasibility and economic viability of 5 concepts forfuture CR systems with supercritical working fluids including storage concepts (some ofthem with a separate low-pressure salt HTF). Reviews and screenings mainly of PCMsfor high temperature applications have been conducted by Hoshi et al. (2005), Kenisarin(2010), Fan and Khodadadi (2011) and Liu et al. (2012). Some properties of the mate-rials, that were decided to be appropriate for high-temperature storage (> 550 ◦C), canbe found in APPENDIX B.1 and B.2.
9. Media for Active Sensible Heat TESSs
9.1. Storage Media in Use
The first two SEGS plants featured a direct TESS with oil as the storage medium.Because of the high cost of the storage medium/HTF, this system has not been buildsince. All other commercial TESSs are based on Solar Salt™ as the storage medium.
24
9.2. Molten Salts
Solar Salt™ (see also Section 3.2) is liquid at ambient pressures between 220 ◦C and600 ◦C which meets the requirements of today’s superheated steam power cycles well.The volumetric heat capacity in this temperature range is good and the specific costof the medium is relatively low (see APPENDIX A.3), however, for state of the arthigh-efficiency supercritical steam or Brayton cycles, the upper temperature poses alimitation. Additionally, the high melting temperature means that at least the pipesand valves of the TESS have to be taken care of in terms of freeze blockage, storage tankinsulation has proven efficient against thermal losses and mechanical stresses. Corrosionof pipes and the tank system can be controlled at a tolerable level (see Kolb, 2011).Experiences of the first commercial-plant-sized salt TESS are summarized in Reilly andKolb (2001). Querol et al. (2012) reported on the construction and tests of a demonstra-tion single-tank active direct TESS with a floating barrier separator between hot andcold part. The 24 MW ht tank has been installed at the Valle2 parabolic trough plant.
To sum up, Solar Salt™ is a good storage medium in many ways. The biggest draw-back is the upper temperature limit which prohibits higher power cycle efficiencies. Therelatively high melting temperature makes freeze protection necessary which, however,mainly influences its use as an HTF. The effect of this is that it has not been used inactive direct TESS for line-focusing CSP application. The medium’s cost is low but, dueto the large amounts of storage medium necessary, plays a big role in CSP plants’ totalcost. Therefore, other media, even if only slightly cheaper, have to be investigated.
Alternative high-temperature or low-melting-point salts (see Section 3.2) are assumedto, at least in the medium term, have much higher specific costs than Solar Salt™. Evenif they are used as an HTF, they will likely not be used as the (sole) storage medium.
Cordaro et al. (2011) presented the results of their measurements of different ther-modynamic properties of several salts and mixtures that are thought to have potentialas TES media. Their measurements show non-linear mixing behavior for properties ofsome of the mixtures, that differ from prior literature.
Zhao and Wu (2011) investigated ternary salt compositions consisting of KNO3,LiNO3 and Ca(NO3)2. Some of the reported solidification temperatures were below100 ◦C, the compositions were stable above 400 ◦C and the costs were only slightly higherthan those of Solar Salt™. The focus of Zhao and Wu was on line-focusing CSP technologyas can be seen from the relatively low temperature stability limit.
9.3. Sodium
Because of its lower density and much higher price as compared to Solar Salt™, sodiumis not viable as a storage medium in the temperature range Solar Salt™ operates in.According to Boerema et al. (2012), the fluid cost and storage volume for sodium ascompared to Solar Salt™ will be 3.2 and 1.8 times higher, respectively. Only when sodiumis used as the HTF and storage medium in a direct configuration or when the highertemperatures are utilized, does its use appear attractive. When technical challenges ofsodium, for example, reactivity with water and possibly corrosion, have been overcome,
25
life-time optimizations of a CSP plant have to compare the benefits in power block andreceiver with the additional risks and cost. Hering et al. (2012) proposed a sodiumthermocline TES working within the temperature span of 200 ◦C and 550 ◦C on theintermediate level of a combined cycle CSP plant (see also Section 3.3.1).
9.4. Haloglass™ RX
Haloglass™ RX is a glass consisting of stable and low-cost components. The upperthermal limit is given by the manufacturer, Halotechnics (2013), as 1200 ◦C, which wouldqualify it as the sensible heat TES medium of the investigated future Rankine or Braytoncycles. However, at 450 ◦C the lower operating limit is high as well and creates freezeprotection challenges. This limit is not dictated by thermal stability or phase change, butby its very high viscosity, which is a typical phenomenon of glasses. The manufacturerstates that at this temperature, the viscosity reaches a pumpability limit of 10 000 mPa s.It is expected that even though technically feasible, operating at these conditions willnot be desirable, so that the upper operating temperature will be considerably higher.At high viscosities, turbulent flow is difficult to reach and, therefore, convective heattransfer is limited.
The thermal conductivity of Haloglass™ RX is mediocre at an estimated value of0.8 W/m K. The influence of radiative heat transfer inside the liquid is assumed to below because of its low transmissivity in the emission spectrum up to 1000 ◦C (Shand,1958). However, this has to be investigated further. More information on this topic canbe found in works by Shand, Turkdogan (1983) and Mann et al. (1992).
The volumetric heat capacity, that is the product of density and specific heat capacity,of Haloglass™ RX is approximately 20 % higher than that of Solar Salt™. Glasses canhave widely varying costs and the price and availability of Haloglass™ RX could notbe identified at this stage. However, due to its proposed application as a “grid scalethermal electricity storage system”, it is assumed to be cost competitive to other TESSsolutions.
9.5. Steam/Water
Steam separators are normal components in (subcritical) steam power cycles. Theyprovide a simple means of separating saturated steam from saturated water in an evap-oration process in order to ensure the steam quality downstream of the evaporator (inthe inlet of the superheater or steam turbine). The mass inside the steam drum alreadyadds some thermal inertia to the system, however, in CSP plant employing DSG, addi-tional inertia is needed for longer cloudy periods or to enable dispatchability of powergeneration.
Abengoa Solar’s Planta Solar 10 (PS10 ) power plant uses four pressurized tanksas so called “steam accumulators”. These store excess thermal energy in times of highirradiation in the form of saturated water, which can be released as steam of continuouslydecreasing pressure (sliding pressure) when needed. This means that the tanks have tostore water at live steam pressure (40 bar in the case of PS10 ), which makes big volumes
26
unviable. At a capacity of 20 MW ht, the vessels of the plant only deliver enough energyto run the turbine at 50 % of the nominal load for 50 minutes (see Solucar, 2006). AbengoaSolar’s second commercial solar power tower plant PS20 has the same storage propertieswhile the nominal power rating of the plant is 20 MWe. The TESS is, therefore, onlya buffer which means that its purpose is to overcome short-term transients. Accordingto Laing et al. (2011), there is no other storage technology commercially available forDSG plants than steam accumulators and these are not cost-competitve for long-termstorage.
Steinmann and Eck (2006) investigated different configurations of steam accumulatorsas TESS. One of them uses a sensible passive concrete heat store in series to the accumu-lator for superheating the saturated steam (Figure 11(a)). Another one (Figure 11(b))features a PCM as filler material in order to enhance the volumetric heat capacity.
Bai and Xu (2011) also investigated a cascaded TESS for DSG plants, consisting ofa steam accumulator and a sensible heat concrete storage. They modeled the thermalbehavior of both TES units during discharging and found a big influence of the thermalconductivity of the sensible storage on the discharging performance.
Saturated Steam
Super heated Steam
Ste
am A
ccum
ulat
or
Sol
id M
ediu
m H
eat
Sto
rage
Saturated Steam
Feedwater
EncapsulatedLatent Heat Storage Material
Figure 11: Enhanced steam accumulators: (a) superheater in series, (b) latent heat fillermaterial (Steinmann and Eck, 2006)
10. Media for Passive Sensible TESS/Filler Material
Solid filler materials are proposed to enhance the thermocline in single-tank TESSs andsubstitute expensive/low-thermal-capacity fluid with solids of higher specific thermalcapacity, higher density and/or lower cost. Popular concepts are packed beds of spheresor natural rocks and high-temperature concrete blocks.
10.1. Packed Beds
Li et al. (2011) and Flueckiger et al. (2013) gave elaborate overviews on heat transfer inpacked beds and its modeling.
27
An unsolved problem of packed bed TESS is an effect called ‘ratcheting’. The termdescribes mechanical stresses on the bed material and on the containment during charg-ing/discharging due to their differing thermal expansion factors (see Figure 12). Dreißi-gacker et al. (2010), Dreißigacker and Zunft (2012) and Dreißigacker et al. (2013) de-scribed the thermo-mechanical modeling and testing of packed bed TESSs. The inves-tigated system was a tank filled with spherical ceramic particles, which was chargedand discharged with unpressurized air of 550 ◦C and 20 ◦C, respectively. They found themechanical stresses on the containment to be “moderate and manageable”.
Figure 12: Rearrangement of particles after several cycles (left) and radial average forcesbefore and during cycling (Dreißigacker and Zunft, 2012).
Spelling et al. (2012) conducted techno-economic analyses of hybridized solar gasturbine plants with and without storage. The TESS was situated downstream of thepressurized receiver and, therefore, pressurized itself. They proposed an insulated steelvessel filled with a packed bed of magnesia fire bricks.
10.2. Rocks and Sand
A packed bed storage built of locally abundant rocks that are virtually cost-free enablesvery cost effective designs. Such storages also have excellent environmental and safetyproperties: They cannot explode, catch fire (except for insulation material perhaps) orleak toxic substances and almost no CO2 is released during manufacturing.
Pacheco et al. (2002) investigated and tested a small pilot-scale thermocline TESwith molten salt as the transport medium and different rock and sand filler materials.
28
They concluded that a combination of quartzite rock and silica sand would be mostappropriate for costs and low voidage fraction because of the different particle sizes. Thisconcept has been used in the Solar One plant before. Brosseau et al. (2005) reported onseveral long-term isothermal (1 year) and thermally cycled (10 000 cycles) tests of thesefiller materials in molten salt environments at temperatures of up to 500 ◦C. The fillermaterials themselves did not show any sign of deterioration, however, at least at thehighest temperatures, the HITEC® XL molten salt did cause extensive corrosion anddeterioration.
Yang and Garimella (2010) simulated the temperature distribution during chargingand discharging in a quartzite rock bed TES with molten salt as the HTF. Theyfound a strong dependence of the tank insulation and Reynolds number on the outlettemperature. Xu et al. (2012) modeled the development of a similar thermocline duringstandby.
Schneider et al. (2011) announced the commercialization of a modular packed bedstorage TESS made of rock or sand. They use ambient air as the secondary HTF ofthe storage system with a heat exchanger separating it from the receiver cycle, whichcould be omitted if air is the primary HTF as well.
Hanchen et al. (2011) developed and validated a thermal model of an air-chargedpacked bed TESS. They investigated the influence of different storage materials (rocks,aluminium, steel, steatite), tank height, mass flow rate and particle diameter.
Zavattoni et al. (2011) and Zanganeh et al. (2012) investigated, modeled and conductedexperiments on a 6.5 MW ht pilot-scale conical packed bed of rocks, which was chargedwith ambient air at temperatures up to 500 ◦C and contained by a concrete structure(see Figure 13). Zavattoni et al. conducted CFD simulations to find pressure drops andtemperature distribution and compared it to experimental results with steatite rockswith an average diameter of 3 cm as the storage medium. The properties of this andother types of storage media that were considered, can be found in APPENDIX B.1.It can be seen, that thermal conductivity of the rocks changes considerably within themeasuring range of 25 ◦C to 175 ◦C.
Allen (2010) and Allen et al. (2012) experimentally investigated pressure drop, heattransfer and thermal cycling stability of packed beds of rock. Of the several differ-ent types of locally (Northern Cape, South Africa) available rock samples, two, namelydolerite and granite, did not show failures when thermally cycled between ambienttemperature and 510 ◦C. They found a strong dependency of the pressure drop - andtherefore the heat transfer - on the packing direction of the rocks. Randomly pouredbeds will, thus, have a high uncertainty in performance. However, the thermal perfor-mance appeared favorable and, based on these experiments, Heller and Gauche (2013)investigated the performance of a rock bed TESS in a combined cycle CSP plant.
Schwaiger et al. (2012) and Haider et al. (2012) investigated and modeled a cascadedsensible/latent TESS for DSG CSP plants. The sensible storage medium of the super-heating section is sand, which is being transported through the heat exchangers viafluidization in air. The aspired maximum storage temperature is approximately 600 ◦C.No details on the properties of the used sand could be retrieved.
Future CR plants with supercritical HTFs (s-H20 and s-CO2) proposed by Kelly (2010)
29
Figure 13: Scheme of the conical rock bed system as set-up by Zanganeh et al. (2012).
feature packed bed thermoclines as the TES. The vessels containing these beds of ce-ramics or quartzite rocks have to withstand the very high pressures of the HTF.According to Kelly, “The most economical pressure vessel is a commercial section ofstandard pipe [. . . ]”. When the largest standard pipes with the necessary wall thicknessare used, thousands of these are needed to enable dispatchability of the plant. Themaximum operating temperature of the TESS is supposed to be approximately 650 ◦C,the particle diameter 5 mm. Capital cost estimates show, that the TESS of the pro-posed supercritical plants are about ten times more expensive than molten salt tanksystems. This is mainly due to the considerably larger amount of high-temperature steelneeded. Even much higher efficiencies in supercritical plants cannot compensate for thishandicap.
10.3. Concrete
Tamme et al. (2003) and Laing et al. (2006) proposed and tested a sensible passive TESSemploying blocks of concrete and castable ceramic, embedding pipes that the HTF passesthrough. The project mainly aimed at the development of a low cost storage due to acheap storage medium for DSG parabolic trough plants. Concrete was found to bethe more fitting material of the two in terms of cost and durability. Improved tubearrangements, distances and enhancements (for example, fins) have been investigated(Laing et al., 2008; Laing, Bahl, Bauer, Fiss, Breidenbach and Hempel, 2012) and, as anadvancement, Laing et al. (2010) investigated a serial arrangement of sensible concreteTESSs for pre- and superheating of steam and a PCM storage for evaporation. Theconcept has been tested with a rating of 1 MW ht as part of a DSG test facility at aconventional power plant (Laing et al., 2011; Laing, Eck, Hempel, Johnson, Steinmann,Meyer and Eickhoff, 2012). The used concrete is thermally stable up to 500 ◦C after
30
several initial heating cycles, in which mass and tensile strength decrease considerably.Further material details of high-temperature concrete and castable ceramics can be foundin APPENDIX B.1.
Brown et al. (2012) and Selvam and Strasser (2012) proposed bricks and parallel platesof concrete in a vessel that are in direct contact to the HTF as the filler material of aTESS. The main advantage of this concept as compared to packed beds is that ratchetingis avoided. The concrete withstood thermal cycling between 300 ◦C and 600 ◦C. Thermalconductivity was given with 2 W/m K, specific heat capacity with 900 J/kg K and thestorage material costs were estimated at 0.78 to 3.18 USD/kW ht.
10.4. Ceramics
Gluck et al. (1991) tested a composite sensible/latent TES consisting of ceramic brickswith molten salt as the PCM. According to Gluck et al., the concept can be explainedas “microencapsulation of a PCM within the submicron pores of a ceramic matrix.”Experiments were run with air of up to 1300 ◦C as the HTF.
As mentioned in Section 10.3, Laing et al. (2006) tested castable ceramic as the storagemedium for parabolic trough DSG plants but found high-temperature concrete moreappropriate for its low price and good thermal stability, despite its lower specific storagecapacity and thermal conductivity.
Dreißigacker et al. (2013) modeled a packed bed built of ceramic spheres for applicationas a TESS for CSP. The focus of their work was on the thermally induced stresses onthe walls and spheres and they didn’t state the type of ceramic they used, but gave someproperties that can be found in B.1.
Zunft, Hanel, Kruger, Dreißigacker, Gohring and Wahl (2011) reported on gained ex-perience from the passive TES subsystem of the Julich Solar Power Tower. “Its TESdesign is based on refractory bricks in honeycomb shape in a stacked arrangement.”(Dreißigacker and Zunft, 2012). The storage is being charged and discharged with un-pressurized air at 680 ◦C and 150 ◦C, respectively. The discharge heat rate reached5.7 MWt at an “almost constant” temperature of 640 ◦C for approximately 1.5 hours.The principle thermal and mechanical performance of the system was confirmed, thepressure drop was lower than expected. In another study, Zunft, Hahn and Kammel(2011) modeled the air flow distribution “[. . . ] in an isotropic porous medium, a wellfounded assumption in particular in the case of ceramic honeycombs or packed bedswith large specific heat transfer surfaces.” They found that the inlet and outlet air dis-tribution system had a considerable influence on flow distribution and, therefore, usefulthermal capacity of the TESS.
10.5. Graphite
Forsberg et al. (2007) proposed graphite as the filler material in the TES of a plantwith high-temperature fluoride salt as the HTF. Graphite is chosen for its compatibilitywith molten salts at high temperatures, its low cost, high heat capacity and thermalconductivity (see APPENDIX B.1). According to the authors, graphite has been used
31
in contact with fluoride salts at temperatures of approximately 1000 ◦C in the aluminiumindustry for decades and the cost of these high performance graphites is less than SunShotcost goals.
11. Latent Heat Storage Media
The determining property of ‘classical’ (eutectic) latent heat TESS (LHTESS) is thatheat is added and rejected at a constant temperature (normally the melting tempera-ture). This ensures constant operating conditions of heat exchangers and turbomachin-ery. However, the discharge rate might be non-constant because of differing heat transfereffects due to solidification of the phase change material (PCM) around heat exchangertubes (see Figure 14). The other aspired advantage of a TESS employing PCMs, besidesthe constant temperature, is the high energy density and therefore smaller mass of thestorage medium and volume of the tank.
di
do
dOB
dMOB
Molten PCM
Solidified PCM
Heat transfer pipe
Figure 14: Scheme of the cross-section of a heat transfer pipe in a latent heat TES(Kotze, Backstrom and Erens, 2012b).
Hoshi et al. (2005) did a screening of potential PCMs for linear Fresnel and CR appli-cations. The most important property is obviously the temperature at which the phasechange happens. Because of the specific volume and required pressure, this is almostexclusively the melting temperature and not the evaporation temperature. In the regionof melting temperatures > 500 ◦C, which is relevant for high-efficiency cycles, all PCMsinvestigated by Hoshi et al. are salts. Figure 15 shows the respective melting pointsof these salts at ambient pressure and another very important property for system costreduction: the volumetric heat capacity. The third important thermophysical propertyof PCMs is thermal conductivity. Low values of it create high thermal resistances duringcharging and discharging and, therefore, limit the possible heat flux of the TESS or re-quire additional heat transfer surfaces (for example, a greater number of pipes or finnedpipes). Besides the qualities given above, Liu et al. (2012) also named the following as
32
important aspects of storage media: chemically stabile, non-corrosive, non-hazardous,congruent melting, insignificant supercooling and low cost.
700 800 900 1000 1100 1200
MeltingFpointFTm
EFK
1300500 600
800
0
400
600
He
atF
ca
pa
city
FEFk
Wh
-m3
200
FluoridesChloridesCarbonatesSulfatesBromidesNitratesHydroxidesEutectics
LiF
NaF
KF
LiClNaCl
KCl
MgCl2
CaCl2
Li2CO3
K2CO3
Na2CO3
Li2SO4 Na2SO4
LiBr
LiNO3
NaNO3KNO3
LiOH
NaOHKOH
NaClLCaCl2
Li2CO3LNa2CO3
LiFLCaF2
Figure 15: Volumetric heat capacity and melting point of investigated salt PCMs (Hoshiet al., 2005).
Conductivity enhancing methods for LHTESS have been investigated and reviewedby Fan and Khodadadi (2011), Agyenim et al. (2010), Liu et al. (2012) and Robak et al.(2011), who proposed heat pipes connected to the HTF pipes. Liu et al., Zalba et al.(2003), Gil et al. (2010) and Kenisarin (2010) also reviewed PCMs in the temperaturerange (see APPENDIX B.2).
11.1. Salts
Laing et al. (2010) describe a combined sensible/latent heat TESS with concrete andNaNO3 as the storage media. The enthalpy of fusion of NaNO3 is 175 kJ/kg at a meltingtemperature of 306 ◦C. This temperature is appropriate for a superheated DSG plantwith charging and discharging pressures of 107 bar and 81 bar, respectively. Detailedthermophysical properties of NaNO3 and its melting behavior can be found in Baueret al. (2009) and APPENDIX B.2. Test results for a pilot combined TESS, which hasbeen installed at a conventional power plant, show that the used finned tubes in the700 kW ht LHTESS improved the discharging rate significantly (Laing, Eck, Hempel,Johnson, Steinmann, Meyer and Eickhoff, 2012). Schwaiger et al. (2012) used the samestorage medium for the latent heat part of their modeled TESS, however, with sandas the storage medium to provide the sensible heat needed for superheating steam (seeSection B.2).
Zipf et al. (2012) developed and tested a screw heat exchanger for the phase changeprocess in a latent heat TESS based on Solar Salt™ as the PCM. The heat exchangeris intended for a DSG cycle at an evaporation temperature of 221 ◦C. According to theauthors, this technology could also be used for other molten salts, for instance, NaNO3,to reach higher evaporation temperatures, pressures and efficiency.
33
11.2. Metals
Blanco-Rodrıguez et al. (2012) screened different metal compositions for the use as PCMsin DSG plants. They decided on the eutectic magnesium-zinc alloy with 49 % by massof magnesium (MgZn51) as the most suitable one and conducted thermo-chemical ex-periences. During 20 freeze-melt cycles, they confirmed literature values between 340 ◦Cand 343 ◦C for the melting temperature and measured the enthalpy of fusion, however,they did not state the result for it.
Kotze, Backstrom and Erens (2012a) investigated the eutectic aluminium-silicon alloywith 12 % by mass of silicon, AlSi12, as the storage medium in between the superheatedsteam power cycle and a molten metal cycle on the receiver side. The chosen moltenmetal of eutectic sodium and potassium (NaK78) reacts highly exothermic with water,so that the chosen TESS also acts as a separating heat exchanger between the twocycles (see Figure 16). The enthalpy of fusion of the eutectic composition is given with549 kJ/kg at a melting temperature of 577 ◦C, which agrees well with state of the artsuperheated steam cycles’ requirements but is low for high efficiency supercritical steamcycles. A possible next generation TESS is proposed to use different metal alloys instead,for example, MgSi56, at a melting temperature of 946 ◦C.
AlSi12 PCM
Housing
Steam/waterpipes
NaK Pipes
Figure 16: AlSi12 heat exchanger/TESS (Kotze, von Backstrom and Ehrens, 2012).
Non-eutectic PCMs do not have a distinct melting temperature but rather a tempera-ture range in which the composition liquifies/solidifies (liquidus to solidus temperature).In order to store energy at a higher temperature and especially if heat is not needed ata constant temperature but over a temperature range, for instance, for superheating orsupercritical heating, non-eutectic compositions can be preferable storage media.
Hunt and Carrington (2012) proposed hypereutectic compositions of aluminium-siliconalloys, that is compositions with a higher mass percentage of silicon than 12.6 % (see
34
Figure 17), as the PCM. The liquidus temperature of hypereutectic Al-Si composi-tions rises up to a maximum of 1414 ◦C for pure silicon, which enables energy storage attemperature levels at the inlet of state of the art gas turbines expanders.
Figure 17: Equilibrium phase change diagram for Al-Si alloys (Hunt and Carrington,2012).
Because of the combined temperature rise and phase change, the effective specific heatcapacity changes during non-eutectic phase change depending on initial composition (seeFigure 18). Heat transfer of hypereutectic Al-Si alloys is good at thermal conductivitiesbetween 190 W/m K for pure silicon and 60 W/m K for AlSi12. The density of Al-Sialloys is relatively high and the density change during melting is small, so that norupturing is to be expected, but measurable, which could support natural conventionwithin heat exchanger tanks. According to Hunt and Carrington (2012), Al-Si alloysare also non-degrading, affordable, available and well known to the industry. The PCMhas to be contained in a material that is able to withstand the high temperatures and(possibly) pressures and doesn’t cause any corrosion issues.
12. Chemical Potential Storage Media
Chemical potential TESS store heat by supplying it to reversible endothermic chemicalreactions. According to Tian and Zhao (2013), the three most important propertiesof the reaction in such a system are: chemical reversibility, large enthalpy change andsimple reaction conditions. Because of the very high energy densities of the reactionproducts, the technology even enables the production of solar fuels, which is being re-searched extensively by the Professorship of Renewable Energy Carriers (2013) at theSwiss Federal Technical University (ETH) Zurich. However, the technology is still in anexperimental to pre-commercial stage, therefore, it is not being reviewed in detail.
35
Figure 18: Changing effective specific heat capacities of Al-Si alloys during melting (Huntand Carrington, 2012).
13. System Considerations
The example of LHTESSs demonstrates the importance of fitting a storage system tothe characteristics of the power block and receiver system. The advantages of PCMsare obvious (high specific capacity and constant stable temperatures). The latter can beimplemented favorably into a saturated/superheated steam power cycle, in which largeamounts of heat are required at a constant temperature for evaporation of the workingfluid. This is even more true if the HTF undergoes a phase change as well, as, forexample, in DSG plants (see Figure 19). However, using a TESS consisting of only onePCM and no sensible storage, would result in high exergetic losses and unnecessary highreceiver inlet temperatures. In most studies, LHTESS have been proposed as one part ofthe storage system for DSG plants (see, for example, Birnbaum et al., 2008; Laing et al.,2010; Feldhoff et al., 2012; Schwaiger et al., 2012). Usually, energy for superheating andpossibly preheating of the working fluid are supplied by sensible heat TESS and only itsheat of evaporation is delivered at constant temperature by the PCM (see Figure 19).
Michels and Pitz-Paal (2007) proposed a cascaded TESS of several PCMs in series forparabolic trough power plants using state of the art oil as the HTF. The temperaturerange of the TESS was therefore approximately 300 ◦C to 400 ◦C. The chosen PCMs withtheir respective melting temperatures and heats of fusion can be found in Figure 20.
An additional application of a (small) LHTESS is the stabilization of the steam gen-erator inlet temperature. In such a system, the PCM would buffer transients throughits high enthalpy of fusion.
Birnbaum et al. (2008) compared two different options of implementing the storagesystem in a DSG plant. In both, the TESS is charged with live steam at nominal 110 barand discharged at 75 bar, which is due to the temperature gradient in the TESS duringcharging and discharging. However, when the solar field only generates enough energyto run the power cycle in part load (see Figure 21), the two options differ: In the first
36
Figure 19: T-s diagram depicting charging and discharging of a cascaded latent andsensible heat TESS (Schwaiger et al., 2012).
one, additional steam mass flow will be added once the live steam pressure reaches thedischarging pressure of the TESS (75 bar), in the second one, steam generated in theTESS only enter the steam turbine after the initial blade rows when the steam from thesolar field is expanded to the pressure level of storage discharge.
Aga et al. (2012) proposed a DSG plant layout in which the turbine-generator systemproduces as much energy during storage discharging mode as in ‘solar’ mode. This isachieved by shutting down the high pressure stage and feeding the intermediate and lowpressure stages of the turbine with higher pressure steam than during nominal load.
Zaversky et al. (2012) compared single-train to parallel dual-train oil-to-salt heat ex-changer configurations for parabolic trough plants. They conclude that the performanceof the parallel concept is much more favorable in part-load because one train could becompletely shut off instead of strongly decreasing the oil mass flow and, therefore, theheat transfer coefficient.
Yogev and Kribus (2012) investigated the discharging behavior of the LHTES partof the latent/sensible hybrid TESS in a CSP plant. They build a simple model ofa LHTES with NaNO3 as the PCM and compared different discharging modes withdifferent heat flux reduction rates. They found that the power cycle’s electricity outputcan be controlled to be nearly constant when it is run in sliding pressure mode and theflow rate is increased accordingly during discharging.
14. Modeling of TESS
Powell and Edgar (2012) described the basic modeling of a CSP plant consisting ofcollector, TESS, boiler and control system. Li et al. (2011) states analytical equations
37
NaNO3
ϑm =w306 °CΔhm =w
172wkJ/kgw
MgCl2/KCl/wNaClw
ϑm =w380 °C wΔhm =
400wkJ/kgw
KOHw
ϑm =w360 °CΔhm =w
134wkJ/kgw
KNO3 /KClw
ϑm =w320w °CwΔhm =w
74wkJ/kgw
KNO3
ϑm =w335w°CΔhm =
95wkJ/kgw
HTF-Flowwwhenwchargingw
HTF-Flowwwhenwdischarging
topw
ofws
tora
ge
botto
mwo
fwsto
rage
Figure 20: Sketch of a cascaded LHTESS including melting temperatures and heats offusion (Michels and Pitz-Paal, 2007).
sola
r fi
eld
outl
et p
ress
ure
standardized power of turbine1
78 bar
min
fixed pressure mode
modified sliding
sliding
110 bar
min
no storagecharge ordischarge
optional storagedischarge
storagecharge
Figure 21: Pressure characteristics of DSG cycle at part-load (Birnbaum et al., 2008).
and numerical approaches for modeling of packed bed passive TESS and LHTESSs. VanLew et al. (2011) compared simulations with the model for packed beds by Li et al.with eperimental results from the literature. The thermo-mechanical modeling of apacked bed TES was demonstrated by Dreißigacker et al. (2010). Zanganeh et al. (2012)modeled a rock bed TESS and validated the model by low-temperature experiments ona 6.5 MW ht demonstration system. Flueckiger et al. (2011b) built a thermo-mechanicalmodel of a TESS that consists of solid filler material (rock and sand) and oil as the activemedium in the storage loop. The validation with data from the Solar One demonstrationplant shows good agreement in predicting thermal ratcheting. In (Flueckiger et al.,2011a), the active medium was changed to molten salt (HITEC® Heat Transfer Salt)and the filler material was only quartzite rock. Xu et al. (2012) modeled a molten saltthermocline tank, varied inlet velocity, porosity, inlet temperature and tank height andshowed the influence on the thermocline quality after various time steps of standby.Yogev and Kribus (2012) modeled an DSG plant including an LHTESS. They simulated
38
the charging and discharging behavior for different thermal conductivity values in slidingpressure mode.
39
References
Aga, V., Ehrsam, A., Boschek, E., Breschi, D. and Girard, R. (2012), Thermal EnergyStorage And Turbine Solution For Direct Steam Solar Power Plants With NominalPower Output During Storage Discharge, in ‘SolarPACES 2012’, Marrakech, Morocco.
Agyenim, F., Hewitt, N., Eames, P. and Smyth, M. (2010), ‘A review of materials, heattransfer and phase change problem formulation for latent heat thermal energy storagesystems (LHTESS)’, Renewable and Sustainable Energy Reviews 14(2), 615–628.URL: http:// linkinghub.elsevier .com/ retrieve/ pii/ S1364032109002469
Allen, K., Backstrom, T. V. and Kroger, D. (2012), Packed Beds of Rock for Ther-mal Storage, in ‘1st Southern African Solar Energy Conference’, Stellenbosch, SouthAfrica, pp. 1–9.
Allen, K. G. (2010), Performance characteristics of packed bed thermal energy storagefor solar thermal power plants, M.sc. eng, University of Stellenbosch.URL: http:// hdl .handle.net/ 10019 .1/ 4329
An, D. (2011), Sodium fire suppression using liquid nitrogen, M.sc., University ofMaryland.URL: http:// drum.lib.umd .edu/ bitstream/ 1903/ 12385/ 1/ An umd 0117N12846 .pdf
Aora Solar Ltd (2012), ‘AORA’.URL: http:// aora-solar .com/
Avila Marın, A. L. (2011), ‘Volumetric receivers in Solar Thermal Power Plants withCentral Receiver System technology: A review’, Solar Energy 85(5), 891–910.URL: http:// www .sciencedirect .com/ science/ article/ pii/ S0038092X11000302
Bai, F. and Xu, C. (2011), ‘Performance analysis of a two-stage thermal energy stor-age system using concrete and steam accumulator’, Applied Thermal Engineering31(1415), 2764–2771.
Bauer, T., Breidenbach, N., Pfleger, N., Laing, D. and Eck, M. (2012), Overview ofMolten Salt Storage Systems and Material Development for Solar Thermal PowerPlants, Technical report, DLR.URL: https:// ases.conference-services.net/ resources/ 252/ 2859/ pdf/ SOLAR20120122 fullpaper .pdf
Bauer, T., Laing, D., Kroner, U. and Tamme, R. (2009), Sodium Nitrate for HighTemperature Latent Heat Storage, in ‘11th International Conference on ThermalEnergy Storage’, Stockholm, Sweden.URL: http:// talon.stockton.edu/ eyos/ energy studies/ content/ docs/ effstock09/Session 3 1 Efficient Soultions using short term TES/ 4 .pdf
40
Becker, M. (1980), ‘Comparison of heat transfer fluids for use in solar thermal powerstations’, Electric Power Systems Research 3(3-4), 139–150.URL: http:// linkinghub.elsevier .com/ retrieve/ pii/ 0378779680900012
Birnbaum, J., Eck, M., Fichtner, M., Hirsch, T., Pitz-Paal, R. and Zimmermann, G.(2008), A Concept for Future Parabolic Trough Based Solar Thermal Power Plants,in ‘ICPWS’, Berlin.URL: www .15icpws.de/ papers/ 10 Green-02 Birnbaum.pdf
Blanco-Rodrıguez, P., Rodrıguez-Aseguinolaza, J., Faik, A., Calvet, N., Tello, M. J.and Doppiu, S. (2012), Eutectic Metal Alloys as Phase Change Material for Ther-mal Energy Storage in Concentrated Solar Power, in ‘SolarPACES 2012’, Marrakech,Morocco.
Boerema, N., Fisher, J., Morrison, G. and Rosengarten, G. (2011), A Comparison ofHigh Temperature Heat Transfer Fluids for Concentrating Solar Thermal Systems, in‘SolarPACES 2011’, Vol. 2, Granada, Spain.
Boerema, N., Morrison, G., Taylor, R. and Rosengarten, G. (2012), ‘Liquid sodiumversus Hitec as a heat transfer fluid in solar thermal central receiver systems’, SolarEnergy 86(9), 2293–2305.URL: http:// www .sciencedirect .com/ science/ article/ pii/ S0038092X12001703
Boss, M., Gadoury, T., Feeny, S. and Montgomery, M. (2007), Recent Advances inUltra Super Critical Steam Turbine Technology, Technical report.URL: http:// www .ge-energy .com/ content/ multimedia/ files/ downloads/ article-GE1000MWUSCpaperwithedits2-1-10 .pdf
Brosseau, D., Kelton, J. W., Ray, D., Edgar, M., Chisman, K. and Emms, B. (2005),‘Testing of thermocline filler materials and molten-salt heat transfer fluids for Thermalenergy storage systems in parabolic trough power plants’, Journal of solar energyengineering 127(1), 109–116.URL: http:// cat .inist .fr/ ?aModele=afficheN&cpsidt=16586587
Brown, B. M., Strasser, M. N. and Selvam, R. P. (2012), Development of a StructuredThermocline Thermal Energy Storage System, pp. 1–8.URL: http:// ases.conference-services.net/ resources/ 252/ 2859/ pdf/ SOLAR20120074 fullpaper .pdf
Cabeza, L. F., Sole, C., Castell, A., Oro, E. and Gil, A. (2012), Review of Solar ThermalStorage Techniques and Associated Heat Transfer Technologies, in ‘Proceedings of theIEEE’, Vol. 100, pp. 525–538.URL: http:// ieeexplore.ieee.org/ xpls/ abs all .jsp?arnumber=5942138&tag=1
Cengel, Y. and Boles, M. (2010), Thermodynamics: An Engineering Approach withStudent Resources DVD, McGraw-Hill Science/Engineering/Math.
41
Chapman, D. J. and Arias, D. A. (2009), An Assessment of the Supercritical CarbonDioxide Cycle for Use in a Solar Parabolic Trough Power Plant, in ‘Proceedings ofSCCO2 Power Cycle Symposium 2009’, Troy, NY.
Chen, H., Chen, Y., Hsieh, H.-T. and Siegel, N. (2007), ‘Computational Fluid DynamicsModeling of Gas-Particle Flow Within a Solid-Particle Solar Receiver’, Journal ofSolar Energy Engineering 129(2), 160.URL: http:// link .aip.org/ link/ JSEEDO/ v129/ i2/ p160/ s1&Agg=doi
Coastal Chemical Co. (n.d.a), ‘HITEC Heat Transfer Salt’.URL: http:// www .coal2nuclear .com/ MSR-HITECHeatTransferSalt .pdf
Coastal Chemical Co. (n.d.b), ‘Hitec Solar Salt’.
Cordaro, J. G., Kruizenga, A. M., Altmaier, R., Sampson, M. and Nissen, A. (2011),Thermodynamic properties of molten nitrate salts, in ‘SolarPACES 2011’, Vol. 1, San-dia National Laboratories, Seville, Spain.URL: http:// energy .sandia.gov/ wp/ wp-content/ gallery/ uploads/ Thermodynamic-Porperties-of-Molten-Nitrate-Salts-Cordaro.pdf
Coventry, J. and Pye, J. (2010), Coupling Supercritical and Superheated Direct SteamGeneration with Thermal Energy Storage, Technical report, Wizard Power.URL: http:// solar-thermal .anu.edu.au/ files/ 2010/ 02/ SolarPACES2009CoventryPye.pdf
Cox, T. L. (2009), Preliminary Results of Heat Transfer Experiments with SupercriticalCarbon Dioxide in the Small System Test Loop, in ‘Proceedings of SCCO2 PowerCycle Symposium 2009’, Troy, NY.
Crocker, A. and Miller, F. (2011), Fluid Flow and Radiation Modeling of a CylindricalSmall Particle Solar Receiver, in ‘SolarPACES 2011’, Granada, Spain.
Diver, R. B., Adkins, D. R. and Moss, T. A. (1990), Trends in Dish-Stirling SolarReceiver Designs, in ‘Energy Conversion Engineering Conference’, pp. 303–310.URL: http:// ieeexplore.ieee.org/ xpls/ abs all .jsp?arnumber=747967
DLR (2010), Solar-Hybrid Power and Cogeneration Plants, Technical Report 019830,Deutsches Zentrum fur Luft- und Raumfahrt e.V.URL: http:// www .fzea.usp.br/ green/ solhyco/ upload/ Solhyco final report .pdf
Dostal, V., Driscoll, M. J. and Hejzlar, P. (2004), A Supercritical Carbon Dioxide Cyclefor Next Generation Nuclear Reactors, Technical report, Massachusetts Institute ofTechnology, Cambridge, MA.
Dow Chemical Company (2001), ‘Synthetic Organic Heat Transfer Fluid Liquid andVapor Phase Data’.URL: http:// www .dow .com/ webapps/ lit/ litorder .asp?filepath=heattrans/ pdfs/noreg/ 176-01463 .pdf&pdf=true
42
Dreißigacker, V., Muller-Steinhagen, H. and Zunft, S. (2010), ‘Thermo-mechanical anal-ysis of packed beds for large-scale storage of high temperature heat’, Heat and MassTransfer 46(10), 1199–1207.URL: http:// www .springerlink .com/ index/ 10 .1007/ s00231-010-0684-5
Dreißigacker, V. and Zunft, S. (2012), Thermo-mechanical implications of packed bedheat storage for CSP : A Modeling approach and a simulation study for a large- scaleapplication, in ‘SolarPACES 2012’, Marrakech, Morocco.
Dreißigacker, V., Zunft, S. and Muller-Steinhagen, H. (2013), ‘A thermo-mechanicalmodel of packed-bed storage and experimental validation’, Applied Energy .URL: http:// linkinghub.elsevier .com/ retrieve/ pii/ S0306261913002638
Duffie, J. A. and Beckman, W. A. (1991), Solar Engineering of Thermal Processes, 2ndedn, John Wiley & Sons, Inc., New York.
Dunn, R. I., Hearps, P. J. and Wright, M. N. (2012), Molten-Salt Power Towers: NewlyCommercial Concentrating Solar Storage, in ‘Proceedings of the IEEE’, Vol. 100,pp. 504–515.
EC (2005), Solar hybrid gas turbine electric power system, Technical report, EuropeanCommission, Brussels.URL: http:// bookshop.europa.eu/ en/ solgate-solar-hybrid-gas-turbine-electric-power-system-pbKINA21615/ downloads/ KI-NA-21-615-EN-C/KINA21615ENC 002 .pdf?FileName=KINA21615ENC 002 .pdf&SKU=KINA21615ENC PDF&CatalogueNumber=KI-NA-21-615-EN-C
Eck, M. and Zarza, E. (2006), ‘Saturated steam process with direct steam generatingparabolic troughs’, Solar Energy 80(11), 1424–1433.URL: http:// www .sciencedirect .com/ science/ article/ pii/ S0038092X06000892
Fan, L. and Khodadadi, J. (2011), ‘Thermal conductivity enhancement of phase changematerials for thermal energy storage: A review’, Renewable and Sustainable EnergyReviews 15(1), 24–46.URL: http:// linkinghub.elsevier .com/ retrieve/ pii/ S1364032110002595
Feldhoff, J. F., Schmitz, K., Eck, M., Schnatbaum-Laumann, L., Laing, D., Ortiz-Vives,F. and Schulte-Fischedick, J. (2012), ‘Comparative system analysis of direct steamgeneration and synthetic oil parabolic trough power plants with integrated thermalstorage’, Solar Energy 86(1), 520–530.URL: http:// www .sciencedirect .com/ science/ article/ pii/ S0038092X11003938
Flueckiger, S. M., Yang, Z. and Garimella, S. V. (2013), ‘Review of Molten-Salt Thermo-cline Tank Modeling for Solar Thermal Energy Storage’, Heat Transfer Engineering34(10), 787–800.URL: http:// dx .doi .org/ 10 .1080/ 01457632 .2012 .746152
43
Flueckiger, S., Yang, Z. and Garimella, S. V. (2011a), ‘An integrated thermal andmechanical investigation of molten-salt thermocline energy storage’, Applied Energy88(6), 2098–2105.URL: http:// www .sciencedirect .com/ science/ article/ pii/ S0306261910005519
Flueckiger, S., Yang, Z. and Garimella, S. V. (2011b), ‘Thermocline Energy Storagein the Solar One Power Plant: An Experimentally Validated Thermomechanical In-vestigation’, Proceedings of the ASME 2011 5th International Conference on EnergySustainability pp. 775–781.URL: http:// dx .doi .org/ 10 .1115/ ES2011-54578
Forsberg, C. W., Peterson, P. F. and Zhao, H. (2007), ‘High-Temperature Liquid-Fluoride-Salt Closed-Brayton-Cycle Solar Power Towers’, Journal of Solar EnergyEngineering 129(2), 141.URL: http:// link .aip.org/ link/ JSEEDO/ v129/ i2/ p141/ s1&Agg=doi
Foust, O. J. (1972), Sodium-NaK Engineering Handbook, Gordon and Breach, SciencePublishers, Inc., New York.
Fuller, R. L. and Batton, W. (2009), Practical Considerations in Scaling SupercriticalCarbon Dioxide Closed Brayton Cycle Power Systems, in ‘Proceedings of SCCO2Power Cycle Symposium 2009’, Troy, NY.URL: http:// www .sco2powercyclesymposium.org/ resource center/ turbomachinery/practical-considerations-in-scaling-supercritical-carbon-dioxide-closed-brayton-cycle-power-systems
Furukawa, T., Muller, G., Schumacher, G., Weisenburger, A., Heinzel, A., Zimmer-mann, F. and Aoto, K. (2004), ‘Corrosion Behavior of FBR Candidate Materials inStagnant Pb-Bi at Elevated Temperature’, Journal of Nuclear Science and Technology41(3), 265–270.URL: http:// www .tandfonline.com/ doi/ abs/ 10 .1080/ 18811248 .2004 .9715484
Garbrecht, O., Al-sibai, F., Kneer, R. and Wieghardt, K. (2012), Numerical Investigationof a New Molten Salt Central Receiver Design, in ‘SolarPACES 2012’, Marrakech,Morocco.
Garcıa, E. and Calvo, R. (2012), One Year Operational Experience of Gemasolar Plant,in ‘SolarPACES 2012’, Marrakech, Morocco.
Gary, J., Turchi, C. and Siegel, N. (2011), CSPand the DOE SunShot Initiative, in‘SolarPACES 2011’, Vol. 1, Granada, Spain.
Gasanaliev, A. M. and Gamataeva, B. Y. (2000), ‘Heat-accumulating properties ofmelts’, Russian Chemical Reviews 69(2), 179–186.URL: http:// stacks.iop.org/ 0036-021X/ 69/ i=2/ a=R07?key=crossref .51b16f738ba53d206a4795cbc793e031
44
Gemasolar (2011), ‘Gemasolar: how it works’.URL: http:// www .torresolenergy .com/ EPORTAL DOCS/ GENERAL/ SENERV2/DOC-cw4cb709fe34477/ GEMASOLARPLANT .pdf
Gil, A., Medrano, M., Martorell, I., Lazaro, A., Dolado, P., Zalba, B. and Cabeza, L. F.(2010), ‘State of the art on high temperature thermal energy storage for power gen-eration. Part 1 - Concepts, materials and modellization’, Renewable and SustainableEnergy Reviews 14(1), 31–55.URL: http:// linkinghub.elsevier .com/ retrieve/ pii/ S1364032109001774
Gluck, A., Tamme, R., Kalfa, H. and Streuber, C. (1991), ‘Investigation of high tem-perature storage materials in a technical scale test facility’, Solar energy materials24(1-4), 240–248.URL: http:// www .sciencedirect .com/ science/ article/ pii/ 016516339190064R
Guidez, J., Martin, L., Chetal, S. C., Chellaparndi, P. and Raj, B. (2008), ‘LessonsLearned from Sodium-Cooled Fast Reactor Operation and Their Ramifications forFuture Reactors with Respect to Enhanced Safety and Reliability’, Nuclear Technology164, 207–220.
Haider, M., Schwaiger, K., Holzleithner, F. and Eisl, R. (2012), ‘A Comparison betweenPassive Regenerative and Active Fluidized Bed Thermal Energy Storage Systems’,Journal of Physics: Conference Series 395, 012053.URL: http:// stacks.iop.org/ 1742-6596/ 395/ i=1/ a=012053?key=crossref .e5a818cb0734d0a7d4d832124d7bcd64
Halotechnics (2012), ‘Saltstream 700 Product Overview’.URL: http:// www .halotechnics.com/ products/ saltstream700 .html
Halotechnics (2013), ‘Haloglass RX’.URL: http:// www .halotechnics.com/ products/ haloglassrx .html
Hanchen, M., Bruckner, S. and Steinfeld, A. (2011), ‘High-temperature thermal storageusing a packed bed of rocks Heat transfer analysis and experimental validation’,Applied Thermal Engineering 31(10), 1798–1806.URL: http:// www .sciencedirect .com/ science/ article/ pii/ S1359431111001062
Heller, L. and Gauche, P. (2013), ‘Modeling of the Rock Bed Thermal Energy StorageSystem of a Combined Cycle Solar Thermal Power Plant in South Africa’, Solar Energy93, 345–356.URL: http:// dx .doi .org/ 10 .1016/ j .solener .2013 .04 .018
Hennecke, K., Schwarzbozl, P., Hoffschmidt, B., Gottsche, J., Koll, G. and Beuter,M. (2009), The Solar Power Tower Julich A solar thermal power plant for test anddemonstration of air receiver technology, in ‘Proceedings of ISES World Congress2007’, pp. 1–4.URL: http:// link .springer .com/ chapter/ 10 .1007/ 978-3-540-75997-3 358?LI=true
45
Hering, W., Stieglitz, R. and Wetzel, T. (2012), ‘Application of liquid metals for solarenergy systems’, EPJ Web of Conferences 33.URL: http:// dx .doi .org/ 10 .1051/ epjconf/ 20123303003
Herrmann, U. and Kearney, D. W. (2002), ‘Survey of Thermal Energy Storage forParabolic Trough Power Plants’, Journal of Solar Energy Engineering 124(2), 145.URL: http:// link .aip.org/ link/ JSEEDO/ v124/ i2/ p145/ s1&Agg=doi
Hischier, I., Hess, D., Lipinski, W., Modest, M. and Steinfeld, A. (2009), ‘Heat TransferAnalysis of a Novel Pressurized Air Receiver for Concentrated Solar Power via Com-bined Cycles’, Journal of Thermal Science and Engineering Applications 1(4), 041002.URL: http:// link .aip.org/ link/ JTSEBV/ v1/ i4/ p041002/ s1&Agg=doi
Ho, C. K. (2010), Sandia Advanced Concepts, Technical Report 505, Sandia NationalLaboratories, Washington D.C.URL: http:// www1 .eere.energy .gov/ solar/ review meeting/ pdfs/ prm2010 snlho.pdf
Hoshi, A., Mills, D. R., Bittar, A. and Saitoh, T. S. (2005), ‘Screening of high meltingpoint phase change materials (PCM) in solar thermal concentrating technology basedon CLFR’, Solar Energy 79(3), 332–339.URL: http:// www .sciencedirect .com/ science/ article/ pii/ S0038092X04003457
Hunt, A. J. and Carrington, K. R. (2012), Metal Latent Heat Thermal Storage System,in ‘SolarPACES 2012’, Marrakech, Morocco.
International Atomic Energy Agency (1999), Status of liquid metal cooled fast reactortechnology, Technical Report April, International Atomic Energy Agency, Vienna.
Jensch, A., Meschke, F. and Uhlig, R. (2012), ‘Pressurized ceramic plate receiver forhigh outlet temperatures’.
Kadoya, K., Matsunaga, N. and Nagashima, A. (1985), ‘Viscosity and thermal conduc-tivity of dry air in the gaseous phase’, Journal of Physical and Chemical ReferenceData 14(4), 947–970.URL: http:// cat .inist .fr/ ?aModele=afficheN&cpsidt=8524491
Kato, Y., Nitawaki, T. and Muto, Y. (2004), ‘Medium temperature carbon dioxide gasturbine reactor’, Nuclear Engineering and Design 230(1-3), 195–207.URL: http:// linkinghub.elsevier .com/ retrieve/ pii/ S0029549303003935
Kearney, D. W. (2001), ‘Engineering Evaluation of a Molten Salt HTF in a ParabolicTrough Solar Field Concept and Project Overview’.URL: http:// www .nrel .gov/ csp/ troughnet/ pdfs/ ulf herrmann salt .pdf
Kelly, B. (2010), Advanced Thermal Storage for Central Receivers with SupercriticalCoolants, Technical report, Abengoa Solar, Lakewood, Colorado.URL: http:// www .osti .gov/ bridge/ product .biblio.jsp?osti id=981926
46
Kenisarin, M. M. (2010), ‘High-temperature phase change materials for thermal energystorage’, Renewable and Sustainable Energy Reviews 14(3), 955–970.URL: http:// linkinghub.elsevier .com/ retrieve/ pii/ S1364032109002731
Kolb, G. J. (2011), An Evaluation of Possible Next-Generation High-TemperatureMolten-Salt Power Towers, Technical Report December, Sandia National Laborato-ries, Albuquerque.URL: http:// prod .sandia.gov/ techlib/ access-control .cgi/ 2011/ 119320 .pdf
Koll, G., Sahraoui, T., Hoffschmidt, B., Khedim, A., Pomp, S., Schwarzbozl, P. andDillig, M. (2011), ALSOL Solar Thermal Tower Power Plant Algeria, in ‘SolarPACES2011’, number august 2009, Granada, Spain.
Kopp, J. E. (2009), Two-Tank Indirect Thermal Storage Designs for Solar ParabolicTrough Power Plants, Msc., University of Nevada, Las Vegas.URL: http:// link .aip.org/ link/ ASMECP/ v2009/ i48906/ p683/ s1&Agg=doi
Kotze, J. P., Backstrom, T. W. V. and Erens, P. J. (2012a), Evaluation of a latentheat thermal energy storage system using AlSi12 as a phase change material, in ‘So-larPACES 2012’, Marrakech, Morocco.
Kotze, J. P., Backstrom, T. W. V. and Erens, P. J. (2012b), NaK as a primary heat trans-fer fluid in thermal solar power installations, in ‘1st Southern African Solar EnergyConference’, Stellenbosch, South Africa.
Kotze, J. P., von Backstrom, T. W. and Ehrens, P. J. (2012), NaK as a Primary HeatTransfer Fluid in Thermal Solar Power Installations, in ‘1st Southern African SolarEnergy Conference’, Stellenbosch, South Africa.
Kulhanek, M. and Dostal, V. (2011), Thermodynamic Analysis and Comparison of Su-percritical Carbon Dioxide Cycles, in ‘Supercritical CO2 Power Cycle Symposium’,Boulder, CO, pp. 2–8.
Laing, D., Bahl, C., Bauer, T., Fiss, M., Breidenbach, N. and Hempel, M. (2012), High-Temperature Solid-Media Thermal Energy Storage for Solar Thermal Power Plants,in ‘Proceedings of the IEEE’, Vol. 100, pp. 516–524.
Laing, D., Bahl, C., Bauer, T., Lehmann, D. and Steinmann, W.-D. (2011), ‘Thermalenergy storage for direct steam generation’, Solar Energy 85(4), 627–633.URL: http:// www .sciencedirect .com/ science/ article/ pii/ S0038092X10002835
Laing, D., Bauer, T., Lehmann, D. and Bahl, C. (2010), ‘Development of a ThermalEnergy Storage System for Parabolic Trough Power Plants With Direct Steam Gen-eration’, Journal of Solar Energy Engineering 132(2), 21011–21018.URL: http:// dx .doi .org/ 10 .1115/ 1 .4001472
Laing, D., Eck, M., Hempel, M., Johnson, M., Steinmann, W.-D., Meyer, M. and Eick-hoff, M. (2012), High Temperature PCM Storage for DSG Solar Thermal Power Plants
47
Tested in Various Operating Modes of Water/Steam Flow, in ‘SolarPACES 2012’,Marrakech, Morocco.
Laing, D., Steinmann, W.-D., Fiss, M., Tamme, R., Brand, T., Bahl, C. and Fiß, M.(2008), ‘Solid Media Thermal Storage Development and Analysis of Modular StorageOperation Concepts for Parabolic Trough Power Plants’, Journal of Solar EnergyEngineering 130(1), 011006.URL: http:// link .aip.org/ link/ JSEEDO/ v130/ i1/ p011006/ s1&Agg=doi
Laing, D., Steinmann, W.-D., Tamme, R. and Richter, C. (2006), ‘Solid media thermalstorage for parabolic trough power plants’, Solar Energy 80(10), 1283–1289.URL: http:// linkinghub.elsevier .com/ retrieve/ pii/ S0038092X06001538
Lata, J. M., Rodriguez, M. and de Lara, M. A. (2008), ‘High Flux Central Receivers ofMolten Salts for the New Generation of Commercial Stand-Alone Solar Power Plants’,Journal of Solar Energy Engineering 130(2), 021002.URL: http:// link .aip.org/ link/ JSEEDO/ v130/ i2/ p021002/ s1&Agg=doi
Lemmon, E. W., McLinden, M. O. and Friend, D. G. (2011), NIST Chemistry Web-Book, in P. J. Linstrom and W. G. Mallard, eds, ‘NIST Standard Reference DatabaseNumber 69’, Gaithersburg, MD.URL: http:// webbook .nist .gov
Li, P. W., Lew, J. V., Karaki, W., Chan, C. L., Stephens, J. and Brien, J. E. O.(2011), Transient Heat Transfer and Energy Transport in Packed Bed Thermal Stor-age Systems, in M. A. D. S. Bernardes, ed., ‘Developments in Heat Transfer’, InTech,chapter 20, pp. 373–416.URL: http:// www .intechopen.com/ books/ developments-in-heat-transfer/ transient-heat-transfer-and-energy-transport-in-packed-bed-thermal-storage-systems
Litwin, R. Z. (2002), Receiver System: Lessons Learned from Solar Two, TechnicalReport March, Sandia National Laboratories, Canoga Park, CA.URL: http:// prod .sandia.gov/ techlib/ access-control .cgi/ 2002/ 020084 .pdf
Liu, M., Saman, W. and Bruno, F. (2012), ‘Review on storage materials and thermalperformance enhancement techniques for high temperature phase change thermal stor-age systems’, Renewable and Sustainable Energy Reviews 16(4), 2118–2132.URL: http:// www .sciencedirect .com/ science/ article/ pii/ S1364032112000214
Ma, Z. and Turchi, C. S. (2011), Advanced Supercritical Carbon Dioxide Power CycleConfigurations for Use in Concentrating Solar Power Systems, in ‘Supercritical CO2Power Cycle Symposium’, number March, Boulder.
Mann, D., Field, R. E. and Viskanta, R. (1992), ‘Determination of specific heat andtrue thermal conductivity of glass from dynamic temperature data’, Warme - undStoffubertragung 27(4), 225–231.URL: http:// dx .doi .org/ 10 .1007/ BF01589920
48
Massidda, L. and Varone, A. (2007), A Numerical Analysis of a High Temperature SolarCollecting Tube, Using Helium as an Heat Transfer Fluid, Technical report, CRS4,Pula, Italy.URL: http:// publications.crs4 .it/ pubdocs/ 2007/ MV07c/ crs4 1180 .pdf
Medrano, M., Gil, A., Martorell, I., Potau, X. and Cabeza, L. F. (2010), ‘State of theart on high-temperature thermal energy storage for power generation. Part 2 - Casestudies’, Renewable and Sustainable Energy Reviews 14(1), 56–72.URL: http:// linkinghub.elsevier .com/ retrieve/ pii/ S1364032109001786
Mertins, M. (2009), Technische und wirtschaftliche Analyse von horizontalen Fresnel-Kollektoren, Dissertation, Technische Universtat Karlsruhe.URL: http:// digbib.ubka.uni-karlsruhe.de/ volltexte/ 1000013884
Michels, H. and Pitz-Paal, R. (2007), ‘Cascaded latent heat storage for parabolic troughsolar power plants’, Solar Energy 81(6), 829–837.URL: http:// linkinghub.elsevier .com/ retrieve/ pii/ S0038092X06002398
Moisseytsev, A. and Sienicki, J. J. (2010), Extension of the Supercritical Carbon DioxideBrayton Cycle for Application to the Very High Temperature Reactor, in ‘Proceedingsof ICAPP ’10’, San Diego, CA.
Moore, R., Vernon, M., Ho, C. K., Siegel, N. P. and Kolb, G. J. (2010), Design Con-siderations for Concentrating Solar Power Tower Systems Employing Molten Salt,Technical Report September, Sandia National Laboratories.URL: prod .sandia.gov/ techlib/ access-control .cgi/ 2010/ 106978 .pdf
Morita, K., Maschek, W., Flad, M., Yamano, H. and Tobita, Y. (2006), ‘ThermophysicalProperties of Lead-Bismuth Eutectic Alloy in Reactor Safety Analyses’, Journal ofNuclear Science and Technology 43(5), 526–536.URL: http:// www .tandfonline.com/ doi/ abs/ 10 .1080/ 18811248 .2006 .9711131
Muto, Y. and Kato, Y. (2007), Optimal Cycle Scheme of Direct Cycle Supercritical CO2 Gas Turbine for Nuclear Power Generation Systems, in ‘International Conference ofPower Engineering 2007’, Hangzhou, China, pp. 86–92.
Novatec Solar (2013), ‘Projects’.URL: http:// novatecsolar .com/
NREL (2013a), ‘Concentrating Solar Power Projects’.URL: http:// www .nrel .gov/ csp/ solarpaces/ by technology .cfm
NREL (2013b), ‘Ivanpah Solar Energy Generating System’.URL: http:// www .nrel .gov/ csp/ solarpaces/ project detail .cfm/ projectID=62
Olson, L. C., Ambrosek, J. W., Sridharan, K., Anderson, M. H. and Allen, T. R.(2009), ‘Materials corrosion in molten LiFNaFKF salt’, Journal of Fluorine Chem-istry 130(1), 67–73.URL: http:// www .sciencedirect .com/ science/ article/ pii/ S002211390800119X
49
Pacheco, J. E. and Dunkin, S. R. (1996), Assessment of Molten-Salt Solar Central-Receiver Freeze-Up and Recovery Events, Technical report, Sandia NationalLaboratories, Albuquerque.URL: http:// www .osti .gov/ bridge/ servlets/ purl/ 202442-mwiLFH/ webviewable/202442 .pdf
Pacheco, J. E., Reilly, H. E. and Gregory, J. (2000), Summary of the solar two test andevaluation program, Technical report, Sandia National Laboratories.URL: http:// www .stratosolar .com/ uploads/ 5/ 6/ 7/ 1/ 5671050/ 23 solar 2 testand evaluation 751185 .pdf
Pacheco, J. E., Showalter, S. K. and Kolb, W. J. (2002), ‘Development of a Molten-SaltThermocline Thermal Storage System for Parabolic Trough Plants’, Journal of SolarEnergy Engineering 124(2), 153–159.URL: http:// dx .doi .org/ 10 .1115/ 1 .1464123
Pacio, J. and Wetzel, T. (2013), ‘Assessment of liquid metal technology status andresearch paths for their use as efficient heat transfer fluids in solar central receiversystems’, Solar Energy 93, 11–22.URL: http:// linkinghub.elsevier .com/ retrieve/ pii/ S0038092X13001357
Pilkington Solar International GmbH (2000), Survey of Thermal Storage for ParabolicTrough Power Plants, Technical Report September, NREL, Golden, CO.
Poplavskii, V. M., Bagdasarov, Y. E., Kamaev, a. a., Tsikunov, A. G., Zybin, V. A.and Ivanenko, V. N. (2004), ‘Danger of Burning of Sodium Coolant’, Atomic Energy96(5), 327–331.URL: http:// link .springer .com/ 10 .1023/ B:ATEN .0000038098 .44375 .1a
Powell, K. M. and Edgar, T. F. (2012), ‘Modeling and control of a solar thermal powerplant with thermal energy storage’, Chemical Engineering Science 71(0), 138–145.URL: http:// www .sciencedirect .com/ science/ article/ pii/ S0009250911008657
Professorship of Renewable Energy Carriers (2013), ‘Publications’.URL: http:// www .pre.ethz .ch/ publications/ journals/
Querol, P., Olano, J., Velasco, T. and Arevalo, J. E. (2012), Single tank thermal storageprototype, in ‘SolarPACES 2012’, Marrakech, Morocco.
Raade, J. W., Elkin, B. and Vaughn, J. (2012), Novel 700 C Molten Salt for SolarThermal Power Generation with Supercritical Steam Turbines, in ‘SolarPACES 2012’,Vol. 001, Marrakech, Morocco.
Raade, J. W. and Padowitz, D. (2011), ‘Development of Molten Salt Heat TransferFluid With Low Melting Point and High Thermal Stability’, Journal of Solar EnergyEngineering 133(3), 31013–31016.URL: http:// dx .doi .org/ 10 .1115/ 1 .4004243
50
Reilly, H. E. and Kolb, G. J. (2001), An Evaluation of Molten-Salt Power Towers Includ-ing Results of the Solar Two Project, Technical report, Sandia National Laboratories.URL: http:// www .osti .gov/ energycitations/ product .biblio.jsp?osti id=791898
Robak, C. W., Bergman, T. L. and Faghri, A. (2011), ‘Economic evaluation of latentheat thermal energy storage using embedded thermosyphons for concentrating solarpower applications’, Solar Energy 85(10), 2461–2473.URL: http:// www .sciencedirect .com/ science/ article/ pii/ S0038092X11002489
Rodrıguez-Sanchez, M. R., Venegas-Bernal, M., Marugan-Cruz, C. and Santana, D.(2013), Thermal, mechanical and hydrodynamic analysis to optimize the design ofmolten salt central receivers of solar tower power plants., in ‘International Conferenceon Renewable Energies and Power Quality’, number 11, Bilbao.URL: http:// www .icrepq .com/ icrepq% 2713/ 238-rodriguez .pdf
Rouillard, F., Charton, F. and Moine, G. (2009), Corrosion behaviour of different metal-lic materials in supercritical CO2 at 550 C and 250 bars, in ‘Proceedings of SCCO2Power Cycle Symposium 2009’, Troy, NY.
Sabharwall, P., Ebner, M., Sharpe, P., Ambrosek, J., Anderson, M., Sridharan, K., Ol-son, L. and Brooks, P. (2010), Molten Salts for High Temperature Reactors : Univer-sity of Wisconsin Molten Salt Corrosion and Flow Loop Experiments Issues Identifiedand Path Forward, Technical Report March, Idaho National Laboratory, Idaho Falls,ID.URL: www .inl .gov/ technicalpublications/ Documents/ 4502649 .pdf
Schiel, W. J. C. and Geyer, M. A. (1988), ‘Testing an external sodium receiver up toheat fluxes of 2.5 MW/m2: Results and conclusions from the IEA-SSPS high flux ex-periment conducted at the central receiver system of the Plataforma Solar de Almeria(Spain)’, Solar Energy 41(3), 255–265.URL: http:// www .sciencedirect .com/ science/ article/ pii/ 0038092X88901430
Schneider, G., Maier, H. and Stenglein, M. (2011), Development of large scale HighTemperature Thermal Energy Storage System ( HTTESS ) for CSP-power plants, in‘SolarPACES 2011’, Granada, Spain.
Schwaiger, K., Haider, M., Walter, H., Puigpinos, M. and Proetsch, A. (2012), ThermalStorage of Superheated Steam in a Combined Sensible/Latent TES, in ‘SolarPACES2012’, Marrakech, Morocco.
Schwarzbozl, P., Buck, R., Sugarmen, C., Ring, A., Marcos Crespo, M. J., Altwegg, P.and Enrile, J. (2006), ‘Solar gas turbine systems: Design, cost and perspectives’, SolarEnergy 80(10), 1231–1240.URL: http:// www .sciencedirect .com/ science/ article/ pii/ S0038092X05003348
Selvam, R. P. and Strasser, M. N. (2012), Development of a Structured ThermoclineThermal Energy Storage System, in ‘World Renewable Energy Forum’, Denver, CO,pp. 1–17.
51
Sengers, J. V. and Watson, J. T. R. (1986), ‘Improved International Formulations forthe Viscosity and Thermal Conductivity of Water Substance’, Journal of Physical andChemical Reference Data 15(4), 1291–1314.URL: http:// www .nist .gov/ data/ PDFfiles/ jpcrd303 .pdf
Shand, E. B. (1958), Glass Engineering Handbook Second Edition, 2nd edn, McGraw-Hill, York, PA.URL: http:// www .amazon.com/ Glass-Engineering-Handbook-Second-Edition/ dp/B000OMIYDO
Siegel, N. P., Bradshaw, R. W., Cordaro, J. B. and Kruizenga, A. M. (2011), ‘Thermo-physical Property Measurement of Nitrate Salt Heat Transfer Fluids’, ASME Confer-ence Proceedings 2011(54686), 439–446.URL: http:// dx .doi .org/ 10 .1115/ ES2011-54058
SIEMENS AG (2010), ‘Steam turbines for CSP plants’.URL: http:// www .siemens.com/ sustainability/ pool/ de/ umweltportfolio/ produkte-loesungen/ energieuebertragung-energieverteilung/ solar-energy steam-turbines.pdf
SIEMENS AG (2012), ‘Industrial Steam Turbines’.URL: http:// www .energy .siemens.com/ hq/ pool/ hq/ power-generation/ steam-turbines/ Industrial Steam Turbines en.pdf
Sienicki, J. J., Moisseytsev, A., Fuller, R. L., Wright, S. A. and Pickard, P. S. (2011),Scale Dependencies of Supercritical Carbon Dioxide Brayton Cycle Technologies andthe Optimal Size for a Next-Step Supercritical CO2 Cycle Demonstration, in ‘SCO2Power Cycle Symposium’, Boulder, CO.URL: http:// www .sco2powercyclesymposium.org/ resource center/ developmentpriorities/ scale-dependencies-of-supercritical-carbon-dioxide-brayton-cycle-technologies-and-the-optimal-size-for-a-next-step-supercritical-co2-cycle-demonstration
Singer, C., Buck, R., Pitz-Paal, R. and Muller-Steinhagen, H. (2010), ‘Assessment ofSolar Power Tower Driven Ultrasupercritical Steam Cycles Applying Tubular CentralReceivers With Varied Heat Transfer Media’, Journal of Solar Energy Engineering132(4).URL: http:// link .aip.org/ link/ JSEEDO/ v132/ i4/ p041010/ s1&Agg=doi
Solucar (2006), Final Technical Progress Report, Technical Report November 2006.
Solugas Consortium (2012), ‘Solugas - Solar Up-scale Gas Turbine System’.URL: http:// www .solugas.com/ index/ index .php
Solutia Inc. (2013), ‘Therminol VP-1 Vapor Phase/Liquid Phase Heat Transfer Fluid’.URL: http:// www .therminol .com/ pages/ products/ eu/ vp-1 .asp
Spelling, J., Guedez, R. and Laumert, B. (2012), The Value of Storage in Hybrid SolarGas-Turbine Power Plants, in ‘SolarPACES 2012’, Marrakech, Morocco.
52
Steinmann, W.-D. and Eck, M. (2006), ‘Buffer storage for direct steam generation’, SolarEnergy 80(10), 1277–1282.URL: http:// linkinghub.elsevier .com/ retrieve/ pii/ S0038092X05002203
Tamme, R., Laing, D. and Steinmann, W.-D. (2003), ‘Advanced Thermal Energy StorageTechnology for Parabolic Trough’, Solar Energy (March), 563–571.URL: http:// link .aip.org/ link/ ASMECP/ v2003/ i36762/ p563/ s1&Agg=doi
Tian, Y. and Zhao, C. Y. (2013), ‘A review of solar collectors and thermal energy storagein solar thermal applications’, Applied Energy 104(0), 538–553.URL: http:// www .sciencedirect .com/ science/ article/ pii/ S0306261912008549
Turchi, C. S. (2009), Supercritical CO2 for Application in Concentrating Solar PowerSystems, in ‘Proceedings of SCCO2 Power Cycle Symposium 2009’, number Figure 1,Troy, NY, pp. 1–5.URL: http:// www .sco2powercyclesymposium.org/ resource center/ system concepts/supercritical-co2-for-application-in-concentrating-solar-power-systems
Turkdogan, E. T. (1983), Physicochemical properties of molten slags and glasses, MetalsSociety, London, UK.URL: http:// books.google.co.za/ books/ about/ Physicochemical properties ofmolten sla.html? id=TzhRAAAAMAAJ&pgis=1
Tyner, C. E., Sutherland, J. P. and Gould, W. R. J. (1995), Solar Two: A Molten SaltPower Tower Demonstration, Technical report, Sandia National Laboratories.URL: large.stanford .edu/ publications/ coal/ references/ docs/ 1605696 .pdf
U.S. Department of Energy (2012a), ‘Advanced Heat Transfer Fluids and Novel ThermalStorage Concepts for CSP Generation’.URL: http:// www1 .eere.energy .gov/ solar/ sunshot/ csp storage awards.html
U.S. Department of Energy (2012b), ‘High Operating Temperature Liquid Metal HeatTransfer Fluids’.URL: http:// www1 .eere.energy .gov/ solar/ sunshot/ muri ucla.html
Van Lew, J. T., Li, P., Chan, C. L., Karaki, W. and Stephens, J. (2011), ‘Analysisof Heat Storage and Delivery of a Thermocline Tank Having Solid Filler Material’,Journal of Solar Energy Engineering 133(2), 2100301–210310.URL: http:// dx .doi .org/ 10 .1115/ 1 .4003685
Wagner, M. J. (2008), Simulation and Predictive Performance Modeling of Utility-ScaleCentral Receiver System Power Plants by, Msc, University of Wisconsin - Madison.URL: http:// sel .me.wisc.edu/ publications/ theses/ wagner08 .zip
Weitzel, P. S. (2011), Steam Generator for Advanced Ultra-Supercritical Power Plants700 to 760C, in ‘ASME 2011 Power Conference’, Denver, CO.URL: www .babcock .com/ library/ pdf/ br-1852 .pdf
53
Williams, D. F. (2006), Assessment of Candidate Molten Salt Coolants for theNGNP/NHI Heat-Transfer Loop, Technical Report June, Oak Ridge NationalLaboratory, Oak Ridge.URL: https:// inlportal .inl .gov/ portal/ server .pt/ gateway/ PTARGS 0 2 3310277 2604 43/ http;/inlpublisher;7087/ publishedcontent/ publish/ communities/inl gov/ about inl/ gen iv technical documents/ ornl tm 2006 69 htl salt .pdf
Wilson Solarpower (2010a), ‘Core Technology’.URL: http:// wilsonsolarpower .com/ index .php/ solution/ coretechnology .html
Wilson Solarpower (2010b), ‘Wilson Heat Exchanger’.URL: http:// wilsonsolarpower .com/ index .php/ nonsolarapps/ 77-wilsonheatexchanger .html
Wright, S. A., Conboy, T. M., Parma, E. J., Tom G. Lewis, Gary A. Rochau and Ahti J.Suo-Anttila (2011), Summary of the Sandia Supercritical CO2 Development Program,in ‘SCO2 Power Cycle Symposium’, Boulder, CO.
Wright, S. A., Conboy, T. M. and Rochau, G. E. (2011), ‘Overview of supercritical CO2power cycle development at sandia national laboratories’.
Wu, W., Gobereit, B., Singer, C., Amsbeck, L. and Pitz-Paal, R. (2011), Direct Absorp-tion Receivers for High Temperatures, in ‘SolarPACES 2011’, Granada, Spain.URL: http:// elib.dlr .de/ 73113/
Xu, C., Wang, Z., He, Y., Li, X. and Bai, F. (2012), ‘Parametric study and standbybehavior of a packed-bed molten salt thermocline thermal storage system’, RenewableEnergy 48(0), 1–9.URL: http:// www .sciencedirect .com/ science/ article/ pii/ S0960148112002583
Yang, M., Yang, X., Yang, X. and Ding, J. (2010), ‘Heat transfer enhancement andperformance of the molten salt receiver of a solar power tower’, Applied Energy87(9), 2808–2811.URL: http:// www .sciencedirect .com/ science/ article/ pii/ S0306261909001834
Yang, Z. and Garimella, S. V. (2010), ‘Molten-salt thermal energy storage inthermoclines under different environmental boundary conditions’, Applied Energy87(11), 3322–3329.URL: http:// www .sciencedirect .com/ science/ article/ pii/ S0306261910001315
Yogev, R. and Kribus, A. (2012), Performance of Solar Thermal Power Plants Operatingfrom PCM Storage, in ‘SolarPACES 2012’, Marrakech, Morocco.
Zalba, B., Marn, J. M., Cabeza, L. F. and Mehling, H. (2003), ‘Review on thermalenergy storage with phase change: materials, heat transfer analysis and applications’,Applied Thermal Engineering 23, 251–283.URL: http:// linkinghub.elsevier .com/ retrieve/ pii/ S1359431102001928
54
Zanganeh, G., Pedretti, A., Zavattoni, S., Barbato, M. and Steinfeld, A. (2012), ‘Packed-bed thermal storage for concentrated solar power Pilot-scale demonstration andindustrial-scale design’, Solar Energy 86(10), 3084–3098.URL: http:// www .sciencedirect .com/ science/ article/ pii/ S0038092X12002812
Zavattoni, S. A., Barbato, M. C., Pedretti, A. and Zanganeh, G. (2011), CFD Simulationof a Pebble Bed Thermal Energy Storage System Accounting for Porosity VariationsEffects, in ‘SolarPACES 2011’, Granada, Spain.
Zaversky, F., Medina, R., Garcıa-barberena, J., Sanchez, M. and Astrain, D. (2012),Part Load Behavior of Oil-to-Molten-Salt Heat Exchanger Configurations for ActiveIndirect Storage Systems in CSP Applications, in ‘SolarPACES 2012’, Marrakech,Morocco.
Zavoico, A. B. (2001), Solar Power Tower Design Basis Document, Technical ReportJuly, Sandia National Laboratories.URL: http:// prod .sandia.gov/ techlib/ access-control .cgi/ 2001/ 012100 .pdf
Zhang, J. and Li, N. (2008), ‘Review of the studies on fundamental issues in LBEcorrosion’, Journal of Nuclear Materials 373(13), 351–377.URL: http:// www .sciencedirect .com/ science/ article/ pii/ S0022311507008422
Zhao, C. Y. and Wu, Z. G. (2011), ‘Thermal property characterization of a low melting-temperature ternary nitrate salt mixture for thermal energy storage systems’, SolarEnergy Materials and Solar Cells 95(12), 3341–3346.URL: http:// www .sciencedirect .com/ science/ article/ pii/ S0927024811004478
Zipf, V., Neuhauser, A., Willert, D., Nitz, P., Gschwander, S. and Platzer, W. (2012),‘High temperature latent heat storage with a screw heat exchanger: Design of proto-type’, Applied Energy (0).URL: http:// www .sciencedirect .com/ science/ article/ pii/ S0306261912008471
Zunft, S., Hahn, J. and Kammel, J. (2011), Flow Distribution Calculations inRegenerator-Type Heat Storage for Solar Tower Plants, in ‘SolarPACES 2011’,Granada, Spain.
Zunft, S., Hanel, M., Kruger, M., Dreißigacker, V., Gohring, F. and Wahl, E. (2011),‘Julich Solar Power Tower - Experimental Evaluation of the Storage Subsystem andPerformance Calculation’, Journal of Solar Energy Engineering 133(3), 31015–31019.URL: http:// link .aip.org/ link/ ?SLE/ 133/ 031019/ 1
55
Appendix
A. Properties of HTFs
A.1. Overview
56
Property Unit SyntheticOil
SolarSalt™
HITEC®
HeatTransferSalt
Low-T Salt(KLiNa/NO2,NO3)
High-TSalt1(LiF-NaF-KF)
High-T Salt2(Saltstream™
700)
Na NaK LBE
Tmin ℃ 15 222 142 75 454 253 97.7 -12.6 125Tmin,pract ℃ 292 290 142 75 500 300 285 285 285Tmax,pract ℃ 393 593 538 550 1000 700 873 785 1670correspondingpressures
bar 11 1–20 n.a. 1 1 1 1 1
ρ kg/m3 815–673 1910–1720
1980–1690
2200–1800
2270–2040
884–745 779–659 10 300–8770
cp kJ/kg K 2.37–2.73 1.49–1.55 1.40–0.95
1.89(@700 ◦C)
1.45(@ 300 ◦C)
1.31–1.27
0.893–(0.872–)0.892
0.146–0.114
λ W/m K 0.0953–0.0771
0.50–0.55 0.44–0.24
0.60–1.00 76.3–49.0
25.5–26.3–24.1
12.5–26.6
µ mPa s 0.25–0.12 3.50–1.03 20.0–1.00
12.4–1.06 8.4(@ 400 ◦C);4.2(@ 500 ◦C)
0.35–0.16
0.279–0.131
1.91–0.725
costa USD/kg 2.10 0.70–0.93 0.80 11.30 2.00 2.00 13.00storage im-plementation
- + + + o o o o o
HTF alsoworking fluid
- - - - - - -/+b - -
afor reference of cost information, see following tablesbAMTEC, see Section 3.3.1
57
Property Unit ambient Air pressurized Air DSG-superheated
s-H2O (USC) s-CO2
Tmin ℃ n.a. n.a. 0 0 n.a.Tmin,pract ℃ 100 100 249 300 32Tmax,pract ℃ 1300 1000 for refer-
ence600 620 850
correspondingpressure(s)
bar 1 20 190 240 78–200
ρ kg/m3 0.934–0.222 18.7–5.47 815–52.0 741–65.1 881–36.1cp kJ/kg K 1.02–1.22 1.02–1.19 4.19–10.4–2.76 (5600–)5.21–
2.86(54.3–)1.28–2.17
λ W/m K 0.0314–0.0966 0.0321–0.0818 0.634–0.0952 0.578–0.10466 0.07–0.103µ mPa s 0.0219–0.0582 0.0221–0.0509 0.110–0.0338 0.0914–0.0351 (0.087–)0.0450–
0.0477costc USD/kg 0 0storage im-plementation
o - - - -
HTF alsoworking fluid
- + + + +
cfor reference of cost information, see following tables
58
A.2. Synthetic Oil Therminol® VP-1/Dowtherm® A
Table 3: Detailed HTF Properties - Synthetic Oil
Property Unit
Tmin (freeze protection) ℃ 15 Dow Chemical Company (2001)lower operating T (ref-erence for below)
℃ 292 Feldhoff et al. (2012)
practical Tmax (refer-ence for below)
℃ 393 Dow Chemical Company (2001)
corresponding pres-sure(s)
bar 11 Dow Chemical Company (2001)
density range ρ kg/m3 815–673 Dow Chemical Company (2001)isobaric specific heat ca-pacity cp
kJ/kg K 2.37–2.73 Dow Chemical Company (2001)
th. Conductivity λ W/m K 0.0953–0.0771 Dow Chemical Company (2001)dyn. Viscosity µ mPa s 0.25–0.12 Dow Chemical Company (2001)cost (2011) USD/kg 2.10 Robak et al. (2011)storage implementation -also working fluid? -comments state of the art
Correlations (Kopp, 2009):
λoil =[0.1381− 0.00008708 t/◦C− 0.0000001729 (t/◦C)2
]W/m K
cp,oil =[1.509 + 0.002496 t/◦C + 0.0000007888 (t/◦C)2
]kJ/kg K
59
A.3. Solar Salt™
Table 4: Detailed HTF Properties - HITEC® Solar Salt™
Property Unit
Tmin (freeze protection) ℃ 222 Coastal Chemical Co. (n.d.b)lower operating T (ref-erence for below)
℃ 290 Gemasolar (2011)
practical Tmax (refer-ence for below)
℃ 593 Coastal Chemical Co. (n.d.b)
corresponding pressures bar 1–20 Kelly (2010)density range ρ kg/m3 1910–1720 Wagner (2008)isobaric specific heat ca-pacity cp
kJ/kg K 1.49–1.55 Wagner (2008)
th. Conductivity λ W/m K 0.50–0.55 Wagner (2008)dyn. Viscosity µ mPa s 3.50–1.03 Wagner (2008)cost (2012) USD/kg 0.50 Pacio and Wetzel (2013)storage implementation +HTF also working fluid -comments
Correlations (Kopp, 2009), based on Wagner (2008):
λsolarsalt =[0.443 + 0.00019 t/◦C
]W/m K
cp,solarsalt =[1.443 + 0.000172 t/◦C)
]kJ/kg K
µsolarsalt =[22.714− 0.12 t/◦C + 0.0002281 (t/◦C)2 − 0.0000001474 (t/◦C)3
]mPa s
60
A.4. HITEC® Heat Transfer Salt
Table 5: Detailed HTF Properties - HITEC® Heat Transfer Salt
Property Unit
Tmin (freeze protection) ℃ 142 Coastal Chemical Co. (n.d.a)lower operating T (ref-erence for below)
℃ 142
practical Tmax (refer-ence for below)
℃ 538 Coastal Chemical Co. (n.d.a)
corresponding pressures bar n.a.density range ρ kg/m3 1980–1690 Coastal Chemical Co. (n.d.a)isobaric specific thermalcapacity cp
kJ/kg K 1.40–0.95 Boerema et al. (2012)
th. Conductivity λ W/m K 0.44–0.24 Coastal Chemical Co. (n.d.a)dyn. Viscosity µ mPa s 20.0–1.00 Coastal Chemical Co. (n.d.a)cost (2001/2002) USD/kg 0.70–0.93 Kearney (2001) Herrmann
and Kearney (2002)storage implementation +HTF also working fluid -comments alternative
to SolarSalt for lin-ear systemsbecause oflower Tmelt
Correlations (Flueckiger et al., 2011a):
ρHITEC = 1938− 0.732 (T/K− 200)
µHITEC = 1000[
exp(− 4.343− 2.0143 (ln(T/K)− 5.011)
)]mPa s
λHITEC =[− 0.000653 (T/K− 260) + 0.4210
]W/m K
61
A.5. Low Temperature Salt
Table 6: Detailed HTF Properties - Low Temperature Salts (KLiNa/NO2,NO3)
Property Unitmidrule Tmin (freezeprotection)
℃ 75 Bauer et al. (2012)
lower operating T (ref-erence for below)
℃ 75 Bauer et al. (2012)
practical Tmax (refer-ence for below)
℃ 550 Bauer et al. (2012)
corresponding pressures bardensity range ρ kg/m3
isobaric specific thermalcapacity cp
kJ/kg K
th. Conductivity λ W/m Kdyn. Viscosity µ mPa scost USD/kgstorage implementation +HTF also working fluid -comments under development
62
A.6. High Temperature Salts
Table 7: Detailed HTF Properties - LiF-NaF-KF
Property Unit
Tmin (freeze protection) ℃ 454 Forsberg et al. (2007)lower operating T (ref-erence for below)
℃ 500 Forsberg et al. (2007)
practical Tmax (refer-ence for below)
℃ 1000 (for reference,Tmelt ≈ 1610)
Forsberg et al. (2007)
corresponding pressures bar 1 Forsberg et al. (2007)density range ρ kg/m3 2200–1800 Williams (2006)isobaric specific thermalcapacity cp
kJ/kg K 1.89 (@ 700 ◦C) Forsberg et al. (2007)
th. Conductivity λ W/m K 0.60–1.00 Forsberg et al. (2007)dyn. Viscosity µ mPa s 12.4–1.06 Forsberg et al. (2007)cost (1971) USD/kg 11.30 Williams (2006)storage implementation oHTF also working fluid -comments under development; HTF cost too high for direct storage
63
Table 8: Detailed HTF Properties - Saltstream™ 700
Property Unit
Tmin (freeze protection) ℃ 253 Raade et al. (2012)lower operating T (ref-erence for below)
℃ 300 Raade et al. (2012)
practical Tmax (refer-ence for below)
℃ 700 Raade et al. (2012)
corresponding pressures bar 1 Raade et al. (2012)density range ρ kg/m3 2270–2040 Halotechnics (2012)isobaric specific thermalcapacity cp
kJ/kg K 1.45 (@ 300 ◦C) Raade et al. (2012)
th. Conductivity λ W/m Kdyn. Viscosity µ mPa s 8.4 (@ 400 ◦C);
4.2 (@ 500 ◦C)Raade et al. (2012)
cost USD/kgstorage implementation oHTF also working fluid -comments Announced in Dec. 2012; HTF cost too high for direct storage
64
A.7. Sodium
Table 9: Detailed HTF Properties - Sodium (Na)
Property Unit
Tmin (freeze protection) ℃ 97.7 Boerema et al. (2012)lower operating T (ref-erence for below)
℃ 285 Boerema et al. (2012)
practical Tmax (refer-ence for below)
℃ 873 Boerema et al. (2012)
corresponding pressures bar 1 Boerema et al. (2012)density range ρ kg/m3 884–745 Boerema et al. (2012)isobaric specific thermalcapacity cp
kJ/kg K 1.31–1.27 Boerema et al. (2012)
th. Conductivity λ W/m K 76.3–49.0 Boerema et al. (2012)dyn. Viscosity µ mPa s 0.35–0.16 Boerema et al. (2012)cost (2012) USD/kg 2.00 Pacio and Wetzel (2013)storage implementation oHTF also working fluid -/ocomments could be used in direct con-
version power cyclesHering et al. (2012)
Correlations (Boerema et al., 2012):
λNa =[124.67− 0.11381 T/K + 5.5226 · 10−5 (T/K)2 − 1.1842 · 10−8 (T/K)3
]W/m K
cp,Na =[1.6582−8.4790 ·10−4 T/K + 4.4541 ·10−7 (T/K)2−2992.6 (T/K)−2
]kJ/kg K
ρNa =[219 + 275.32 (1− T/2503.7 K) + 511.58 (1− T/2503.7 K)0.5
]kg/m3
µNa =[1000 exp(−6.4406− 0.3958 log(T/K) + 556.835 K/T )
]mPa s
Correlations (Foust, 1972):
λNa =[91.8− 0.049 t/◦C
]W/m K
ρNa = 1000[0.9501−2.2976·10−4 t/◦C−1.46·10−8 (t/◦C)2+5.638·10−12 (t/◦C)3
]kg/m3
µNa =
[0.1235 (ρNa/1000 kg
m3 )1/3 exp(697 (ρNa/1000 kg
m3 ) ·K/T)]
mPa s if T ≤ 500 ◦C,[0.0851 (ρNa/1000 kg
m3 )1/3 exp(1040 (ρNa/1000 kg
m3 ) ·K/T)]
mPa s if T > 500 ◦C,
65
A.8. Eutectic Sodium-Potassium Alloy
Table 10: Detailed HTF Properties - Eutectic Sodium-Potassium Alloy (NaK-78)
Property Unit
Tmin (freeze protection) ℃ -12.6 Foust (1972)lower operating T (ref-erence for below)
℃ 285 Boerema et al. (2012)
practical Tmax (refer-ence for below)
℃ 785 Foust (1972)
corresponding pressures bar 1 Foust (1972)density range ρ kg/m3 779–659 Foust (1972)isobaric specific thermalcapacity cp
kJ/kg K 0.893–0.872–0.892 Foust (1972)
th. Conductivity λ W/m K 25.5–26.3–24.1 Foust (1972)dyn. Viscosity µ mPa s 0.279–0.131 Foust (1972)cost (2012) USD/kg 2.00 Pacio and Wetzel (2013)storage implementation oHTF also working fluid -comments
Correlations (Foust, 1972):
λNaK =[21.4 + 0.0207 t/◦C− 2.2 · 10−5 (t/◦C)2
]W/m K
ρNaK = 1/vNaK
vNaK = 1.003 (0.778 vK + 0.222 vNa)
vK =[0.001/
(0.8415−2.172·10−4 t/◦C−2.70·10−8 (t/◦C)2+4.77·10−12 (t/◦C)3
)]m3/kg
vNa = 1/vNa (see Section A.7)
cp,NaK = 4.184[0.2320− 8.82 · 10−5 (t/◦C) + 8.2 · 10−8 (t/◦C)2
]kJ/kg K
µNaK =
[0.116 (ρNaK/1000 kg
m3 )1/3 exp(688 (ρNaK/1000 kg
m3 ) ·K/T)]
mPa s if T ≤ 400 ◦C,[0.082 (ρNaK/1000 kg
m3 )1/3 exp(979 (ρNaK/1000 kg
m3 ) ·K/T)]
mPa s if T > 400 ◦C,
66
A.9. Eutectic Lead-Bismuth Alloy
Table 11: Detailed HTF Properties - Eutectic Lead-Bismuth Alloy (LBE)
Property Unit
Tmin (freeze protection) ℃ 125 Morita et al. (2006)lower operating T (ref-erence for below)
℃ 285 Boerema et al. (2012)
practical Tmax (refer-ence for below)
℃ 1670 Morita et al. (2006)
corresponding pressures bar 1 Morita et al. (2006)density range ρ kg/m3 10300–8770 Morita et al. (2006)isobaric specific thermalcapacity cp
kJ/kg K 0.146–0.114 Morita et al. (2006)
th. Conductivity λ W/m K 12.5–26.6 Morita et al. (2006)dyn. Viscosity µ mPa s 1.91–0.725 Morita et al. (2006)cost (2012) USD/kg 13.00 Pacio and Wetzel (2013)storage implementation oHTF also working fluid -comments
Correlations (Morita et al., 2006):
λLBE =[6.854 + 1.018 · 10−2 T/K
]W/m K
ρLBE =[10981.7− 1136.9 · 10−3 T/K
]kg/m3
µLBE = 1000[0.49 · 10−3 exp(760.1K/T )
]mPa s for 398 K < T < 1273 K
cp,NaK =[(159− 2.302 · 10−2 T/K)/1000
]kJ/kg K for 400 K < T < 1100 K
67
A.10. Ambient Air
Table 12: Detailed HTF Properties - Ambient Air
Property Unit
Tmin (freeze protection) ℃ n.a.lower operating T (ref-erence for below)
℃ 100
practical Tmax (refer-ence for below)
℃ n.a. (1300 for reference)
corresponding pressures bar 1density range ρ kg/m3 0.934–0.222 ideal gasisobaric specific thermalcapacity cp
kJ/kg K 1.02–1.22 Cengel and Boles (2010)
th. Conductivity λ W/m K 0.0314–0.0966 Kadoya et al. (1985)dyn. Viscosity µ mPa s 0.0219–0.0582 Kadoya et al. (1985)cost USD/kg 0storage implementation oHTF also working fluid -comments advantage of open cy-
cle: HTF impurities, e.g.through rock bed storage,not an issue
Correlations:
ρair = pairRairTair
Ideal gas law
cp,air =[(0.9703+6.7898·10−5(T/K)+1.6576·10−7(T/K)2−6.7863·10−11(T/K)3
]kJ/kg K
(Cengel and Boles, 2010)
λair = 0.0259778[C1Tr +C0.5T
0.5r +
−4∑i=0
CiTir +
5∑j=1
Djρjr
]W/m K (Kadoya et al., 1985)
with Tr = T/132.5 K; ρr = ρ/314.3 kg/m3;C1 = 0.239503;C0.5 = 0.00649768;C0 = 1;C−1 = −1.92615;C−2 = 2.00383;C−3 = −1.07553;C−4 = 0.229414;D1 = 0.402287;D2 = 0.356603;D3 = −0.163159;D4 = 0.138059;D5 = −0.0201725
µair = 0.0061609[A1Tr +A0.5T
0.5r +
−4∑i=0
AiTir +
5∑j=1
Bjρjr
]mPa s (Kadoya et al., 1985)
with Tr = T/132.5 K; ρr = ρ/314.3 kg/m3;C1 = 0.239503;A1 = 0.128517;A0.5 = 2.60661;A0 = −1;A−1 = −0.709661;A−2 = 0.662534;A−3 = −0.197846;A−4 = 0.00770147;B1 = 0.465601;B2 = 1.26469;B3 = −0.511425;B4 = 0.274600
68
A.11. Pressurized Air
Table 13: Detailed HTF Properties - Pressurized Air
Property Unit
Tmin (freeze protection) ℃ n.a.lower operating T (ref-erence for below)
℃ 100
practical Tmax (refer-ence for below)
℃ n.a. (1000 for reference) EC (2005)
corresponding pressures bar 20density range ρ kg/m3 18.7–5.47 ideal gasisobaric specific thermalcapacity cp
kJ/kg K 1.02–1.19 Cengel and Boles (2010)
th. Conductivity λ W/m K 0.0321–0.0818 Kadoya et al. (1985)dyn. Viscosity µ mPa s 0.0221–0.0509 Kadoya et al. (1985)cost USD/kg 0storage implementation -HTF also working fluid +comments advantage of open cy-
cle: HTF impurities (e.g.,through rock bed storage)not an issue
Correlations: see Section A.10
69
A.12. DSG - Superheated
Table 14: Detailed HTF Properties - Superheated Direct Steam Generation
Property Unit
Tmin (freeze protection) ℃ 0lower operating T (ref-erence for below)
℃ 249 NREL (2013b)
practical Tmax (refer-ence for below)
℃ 600 SIEMENS AG (2012)
corresponding pressures bar 190 SIEMENS AG (2012)density range ρ kg/m3 815–52.0 Sengers and Watson (1986)isobaric specific thermalcapacity cp
kJ/kg K 4.19–10.4–2.76 Cengel and Boles (2010)
th. Conductivity λ W/m K 0.634–0.0952 Sengers and Watson (1986)dyn. Viscosity µ mPa s 0.110–0.0338 Sengers and Watson (1986)cost USD/kg Pacio and Wetzel (2013)storage implementation -HTF also working fluid +comments High heat of evaporation
70
A.13. Supercritical H2O
Table 15: Detailed HTF Properties - Ultra-Supercritical Direct Steam Generation
Property Unit
Tmin (freeze protection) ℃ 0lower operating T (ref-erence for below)
℃ 300
practical Tmax (refer-ence for below)
℃ 620 Singer et al. (2010)
corresponding pressures bar 240 Singer et al. (2010)density range ρ kg/m3 741–65.1 Lemmon et al. (2011)isobaric specific thermalcapacity cp
kJ/kg K (5600–)5.21–2.86 Lemmon et al. (2011)
th. Conductivity λ W/m K 0.57884–0.10466 Lemmon et al. (2011)dyn. Viscosity µ mPa s 0.0914–0.0351 Lemmon et al. (2011)cost USD/kg Pacio and Wetzel (2013)storage implementation -HTF also working fluid +comments extrema around critical
point
71
A.14. Supercritical CO2
Table 16: Detailed HTF Properties - supercritical CO2 (s-CO2)
Property Unit
Tmin (freeze protection) ℃ n.a.lower operating T (ref-erence for below)
℃ 32 Moisseytsev and Sienicki (2010)
practical Tmax (refer-ence for below)
℃ n.a. (850 for reference) Moisseytsev and Sienicki (2010)
corresponding pressures bar 78–200 Moisseytsev and Sienicki (2010)density range ρ kg/m3 881–36.1 Lemmon et al. (2011)isobaric specific thermalcapacity cp
kJ/kg K (54.3–)1.28–2.17 Lemmon et al. (2011)
th. Conductivity λ W/m K 0.0779–0.103 Lemmon et al. (2011)dyn. Viscosity µ mPa s (0.087–)0.0450–0.0477 Lemmon et al. (2011)cost USD/kgstorage implementation -HTF also working fluid +comments low compression work
around critical point; highdensity; extrema aroundcritical point
Correlations:
cp,CO2 =[(0.5058+1.359·10−3(T/K)+7.955·10−7(T/K)2−1.697·10−10(T/K)3
]kJ/kg K
(Cengel and Boles, 2010)
72
B. Storage Media
NB: Most of the references are secondary sources.
B.1. Potential Sensible Heat Storage Media
73
Material/Composition(mass-%)
Tlow[℃] Thigh[℃] ρ [kg/m3] cp [kJ/kg K] λ [W/m K] price [USD/kg] Ref.
Liquids
Solar Salt™ 290 593 1790 1.49–1.54 0.50–0.55 0.64 a b
NaNO3 306 700(?) 1900 0.514 c
LiF-NaF-KF 454 ≈ 1610 2200–1800 1.89d 0.006–0.01 64.06 e f
Saltstream™ 700 253 700 2270–2040 1.45g 8.4 h; 4.2i j k
Na 97.7 873 884–745 1.27–1.31 0.16–0.35 2.00 l m
NaK78 -12.6 785 779–659 0.872–0.893 24.1–26.3 2.00 n m
Pb-Bi (44.5-55.5) 125 1670 10300–8770 0.146–0.114 12.5–26.6 42.23 f o
HaloglassTM RX 450 1200 2400 1.36 0.8 p
aCoastal Chemical Co. (n.d.b), Gemasolar (2011), Kopp (2009)bHerrmann and Kearney (2002)c(Yogev and Kribus, 2012; Bauer et al., 2009)d@ 700 ◦CeForsberg et al. (2007)fWilliams (2006)g@ 300 ◦Ch@ 400 ◦Ci@ 500 ◦Cj“too high for direct storage”kRaade et al. (2012) Halotechnics (2012)lBoerema et al. (2012)
mPacio and Wetzel (2013)nFoust (1972)oMorita et al. (2006)pHalotechnics (2013)
74
Material/Composition(mass-%)
Tlow[℃] Thigh[℃] ρ [kg/m3] cp [kJ/kg K] λ [W/m K] price [USD/kg] Ref.
Solids‘N4’ High-TemperatureConcrete
500 2250a 1.10 b 1.3 b c
High-Temperature Con-crete
2680 900 2.0 0.78–3.18 /kW htd
Solid NaCl 500 2160 850 7.0 0.15 e
Silica Fire Bricks 700 1820 1.00 1.5 1.0 e
Magnesia Fire Bricks 1200 3000 1.15 5.0 2.00 e
Ceramics ≥ 700 2400 0.85 1.3 f
Graphite 1700 1.900 200 g
RocksQuartzite 2618 0.623 5.39–3.37h i
Calcareous sandstone 2661 0.652 4.36–2.98i i
Helvetic siliceous lime-stone
2776 0.669 3.60–2.72i i
Limestone 2697 0.683 2.82–2.05i i
Gabbro 2911 0.643 2.05 i
Quartzite Rock 2201 0.964 j
a@ 370 ◦CbLaing, Bahl, Bauer, Fiss, Breidenbach and Hempel (2012)cLaing et al. (2008)dSelvam and Strasser (2012)ePilkington Solar International GmbH (2000)fDreißigacker et al. (2013)gForsberg et al. (2007)h@ 25–175 ◦CiZanganeh et al. (2012)jFlueckiger et al. (2011a)
75
Material/Composition(mass-%)
Tlow[℃] Thigh[℃] ρ [kg/m3] cp [kJ/kg K] λ [W/m K] price [USD/kg] Ref.
Steatite (magnesiumsilicate rock)
2680 1.068 2.5 k
Granite 2893 0.845l 3 m
Dolerite 2657 0.839l 3 m
kHanchen et al. (2011)l@45 ℃
mAllen (2010)
76
B.1.1. Haloglass™ RX
Table 19: Some properties of HaloglassTM RX (Halotechnics, 2013).
Property Value
Melting Point 450 ◦CMaximum Operating Temperature 1200 ◦CDensity 2400 kg/m3
cp at 450 ◦C 1.362 kJ/kg Kλ 0.8 W/m Kµ at 450 ◦C 10 064 mPa s
600 ◦C 600 mPa s800 ◦C 84.3 mPa s1000 ◦C 23.6 mPa s1200 ◦C 11.1 mPa s
77
B.2. PCMs with Tmelt > 550 ◦C
78
Composition (mass-%) Tmelt[℃] ∆hf ρ [kg/m3] cp [kJ/kg K] λ [W/m K] price Ref.[kJ/kg] solid liquid solid liquid solid liquid [USD/kg]
SaltsBaCl2-CaCl-KCl(47-29-24)
551 219 2930 0.67 0.84 0.95 c
CaCl2-BaCl2-KCl(47-29-24)a
551 219 2930 0.67 0.84 0.95 0.20 b
CaCl-CaSO4-CaMoO4(38.5-11-4[?])a
673 224 b
CaF2-CaSO4-CaMoO4(49-41.4-9.6)a
943 237 b
Ca(NO3)2 560 145 c
KBr-KF(60-40)a 576 315 b
KBr-K2MoO4(65-35)a 625 90.5 b
K2CO3 897 236 2290 2.0 h
K2CO3-Li2CO3-Na2CO3(62-22-16)
580 288 2340 1.80 2.09 1.95 0.66 b
K2CO3-Na2CO3(50-50) 710 163 b
KCl 771 353 c
KCl-NaCl-NaF(50.2-39.4-10.4)
602 370.3/m3i i
KF 857 452 2370(?) e
KF-CaF2(85-15)a 780 440 b
KF-KCl(55-45)a 605 407 b
KF-MgF2(85-15)a 790 520 b
KF-NaF-MgF2(63.8-27.9-8.3)
685 2090 e
amol.-%cLiu et al. (2012)bKenisarin (2010)
79
Composition (mass-%) Tmelt[℃] ∆hf ρ [kg/m3] cp [kJ/kg K] λ [W/m K] price Ref.[kJ/kg] solid liquid solid liquid solid liquid [USD/kg]
LiCl-MgF2(94.5-5.5)a 573 131 b
LiF 850 b
Li2CO3-LiF(75.2-24.8) 595 594.5 i
LiF-CaF2(80.5-19.5)a 767 790–820 2100 2670 1.97 1.77–1.84 1.70–3.8 1.70–5.9 b c d g
LiF-CeF3(80-20)a 756 500 b
LiF-KF-MgF2(74-13-13)a
749 860 b
LiF-MgF2(70-30)a 728 520 b
LiF-MgF2(67-33) 746 947 2630(?) 1.42 4.66 b
LiF-MgF2-KF(64-30-6)a 710 782 c
LiF-NaF(60-40)a 652 816 b
LiF-NaF-CaF2(52-35-13)a
615 640 b
LiF-NaF-MgF2(62-19-19)a
693 690 b
LiF-NaF-MgF2(46-44-10)a
632 858 b
Li2SO4-CaSO4-CaMoO4(82-11.44-6.56)a
680 207 b
MgCl2 714 452 2140 e
MgF2-KF(70.5-29.5) 1006 770/m3i i
MgF2-LiF(54.2-45.8) 746 847/m3i 2880 2305 e
MgF2-LiF-CaF2-NaF(37.25–.37.6-34.51–34.79-24.5–25.0-3.21–3.31)a
651–657 460–470 b
iGasanaliev and Gamataeva (2000)dAgyenim et al. (2010)
80
Composition (mass-%) Tmelt[℃] ∆hf ρ [kg/m3] cp [kJ/kg K] λ [W/m K] price Ref.[kJ/kg] solid liquid solid liquid solid liquid [USD/kg]
MgF2-NaF(69-31) 996 710/m3i i
NaBr-NaF(73-27)a 642 360 b
NaCl 800 467–492 2160 5.0 c e g
NaCl-LiF(76-24) 680 476.9/m3i i
NaCl-NaF(66.5-33.5)a 675 572 b
NaCl-Na2MoO4-NaBr(38.5-38.5-23)a
612 168 b
NaCl-NiCl2(52-48)a 573 558 b
NaF 1000 0.74 b
NaF-CaF2(68-32)a 810 600 b
NaF-CaF2-LiF-MgF2(36.5-27.2-25.7-10.6)
593 510/m3 i
NaF-CaF2-MgF2(65-23-12)a
745 568–574 1580(?) 1.17 b
NaF-CaF2-MgF2(51.8-34.0-14.2)
645 2970 2370 e
NaF-LiF(51.9-48.1) 652 711/m3i 2720 1930 e
NaF-LiF-CaF2(38.3-35.2-26.5)a
615 636 2820j b e
NaF-LiF-MgF2(50.4-32.6-17)a
622–632 625/m3i 2810 2105 i e
NaF2-LiF-MgF2(49.9-33.4-17.1)
650 860 2820(?) 1.42 1.15 b
NaF-MgF2(75-25) 832 627–650 2680(?) 1.42 4.66 b
NaF-MgF2(66.9-33.1) 832 2940 2190 e
j@ 20 ◦CeZalba et al. (2003)
81
Composition (mass-%) Tmelt[℃] ∆hf ρ [kg/m3] cp [kJ/kg K] λ [W/m K] price Ref.[kJ/kg] solid liquid solid liquid solid liquid [USD/kg]
NaF-MgF2-KF(64-20-16)a
804 650 b
NaF-MgF2-KF(62.5-22.5-15)a
809 543 b
NaF-MgF2-KF(53.6-28.6-17.8)
809 2850 2110 e
Na2CO3 854 276 2533 2.0 2.6 h
NaCO3-BaCO3/MgO 500–850 420 2600 5.0 2.0 h
Na2SO4 884 165 c
Metals and metal al-loys
b
Al 660 397 c
Al-Si(12.24-87.76) 576 460–560 2540–2700(?) 1.038 1.741 160(?) 190 2.2 b c f
Al-Si(20.0-80.0) 585 460 g
Al-Si(≥12.6-≤87.4) 577–1414 2400–2700 60–190(?) 2.2 f g
Al-Si-Cu(46.3-4.6-49.1) 571 406 5560 c
Al-Si-Cu(65-5-30) 571 422 2730 1.30 1.20 c
Al-Si-Sb(86.4-9.6-4.2) 575 471 2700 c
Al-Si-Mg(83.14-11.7-5.16)
555 485 2500 c
Cu 1083 193.4 c
Cu-Si(80-20) 803 197 6600 0.50 b c
Cu-Si-Mg(56-27-17) 770 420 4150 0.75 b c
Cu-P-Si(83-10-7) 840 92 6880 c b
Cu-P(91-9) 715 134 5600 c b
hPilkington Solar International GmbH (2000)fHunt and Carrington (2012)gGil et al. (2010)
82
Composition (mass-%) Tmelt[℃] ∆hf ρ [kg/m3] cp [kJ/kg K] λ [W/m K] price Ref.[kJ/kg] solid liquid solid liquid solid liquid [USD/kg]
Cu-Zn-P(69-17-14) 720 368 7000 b c
Cu-Zn-Si(74-19-7) 765 125 7170 b c
Mg-Ca(84-16) 790 272 1380 b c
Mg2Cu 841 243 b
Mg-Si-Zn(47-38-15) 800 314 b c
Si-Mg(56-44) 946 757 1900 0.79 b c
Si-Mg-Ca(49-30-21) 865 305 2250 b c
Zn-Cu-Mg(49-45-6) 703 176 8670 0.42 c
Zn2Mg 588 230 b
83